Description:FIELD OF THE INVENTION

The present disclosure relates to thefield of architectural design and computer-aided design, specifically, a system and method for visualizing and generating residential floor plans. The invention focuses on establishing end-to-end pipelines capable of generating conceptual floor plans, conceptual massing models, and realistic visualizations.

BACKGROUND OF THE INVENTION

In the architectural design domain, the prevailing process relies heavily on conventional Computer-Aided Design (CAD) tools like AutoCAD. This method entails architects investing substantial time and effort in manually crafting floor plans, engaging in a repetitive and time-consuming trial-and-error loop to arrive at a single design solution. Yet, this method faces several challenges that disrupt the architects' creativity and efficiency. Architects traditionally invest significant effort into manually drafting floor plans, diverting their attention from other crucial creative aspects. To better understand circulation, functionality, and spatial efficiency, architects often over-detail conceptual floor plans which in turn consumes valuable resources and limits architects' ability to explore broader design concepts.
Additionally, constrained by time and the iterative nature of the process, architects tend to explore only a handful of design variations. This restriction stifles creativity and overlooks potentially innovative solutions lying beyond the current scope of exploration. Moreover, architects often lean on personal design inclinations, potentially bypassing more optimal solutions for common design challenges. This subjective bias can result in missed opportunities for innovative problem-solving.
In essence, the conventional architectural design process struggles with inefficiency, restricted design exploration, and potential biases. Manual labour for floor plan creation and the repetitive nature of the process prevents architects from fully leveraging their creative abilities.
To address these challenges and elevate the architectural design process, the invention introduces a transformative approach. By harnessing evolutionary techniques, 3D projection, and image generation diffusion models, the invention automates key aspects of design. This liberation from manual tasks enables architects to delve deeper into innovative design thinking, explore a wider array of solutions, and make informed decisions based on data-driven insights.
This invention allows architects to allocate more time to creative thinking, diverse solution exploration, and optimized design realization. The integration of this new pipeline breaks free from the limitations of the past and It empowers architects to emphasize core creative pursuits while using data-driven insights to improve their design decisions.
In view of the foregoing discussion, it is portrayed that there is a need to have a system and method for generating a floor plan.

SUMMARY OF THE INVENTION

The present disclosure seeks to provide a system and method for visualizing and generating residential floor plansthat encompasses enhancing architects' design workflow quality and efficiency, minimizing time consumption and human errors, and ensuring easy integration with the existing design pipeline to create and visualize the conceptual design.

In an embodiment,a system for generating a floor planis disclosed. The system includes a data input unit configured to receive architectural data and client preferences for room layouts and plot boundaries.
The system further includes a processing unit coupled to the data input unit, the processing unit configured to encapsulate the received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements, and initialize each room object by selecting a random value from an area range and ratio denominator range, and calculate dimensions based on the selected values.
The system further includes an evolutionary interface connected to the processing unit, the evolutionary interface configured to: i) generate sets of rooms in each iteration, each set representing potential room combinations as individual solutions; ii) evaluate the solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints; and iii) employ custom routines for weighted selection, crossover, and mutation of the solutions to improve fitness scores.
The system further includes a graphical user interface connected to the evolutionary interface, configured to allow users to select one or more of the generated solutions, view each solution alongside its respective fitness score, and input design style preferences, upon which the system generates multiple realistic views.
The system further includes a transformation processing unit connected to the graphical user interface, configured to transform selected floor plans into 3D massing diagrams.
The system further includes a controlling unit in continuation with the transformation processing unit, configured to produce realistic images from the 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance a realism of generated images.
The system further includes a display unit coupled to the controlling unit, configured to export a final 2D floor plan.

In an embodiment, a method for generating a floor plan is disclosed. The method includes receiving architectural data and client preferences for room layouts and plot boundaries through a data input unit.
The method further includes processing received architectural data and client preferences using a processing unit by encapsulating the received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements; and initializing each room object by selecting a random value from an area range and ratio denominator range, and calculating dimensions based on the selected values.
The method further includes generating sets of rooms in each iteration, each set representing potential room combinations as individual solutions using an evolutionary interface and evaluating the solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints thereby employing custom routines for weighted selection, crossover, and mutation of the solutions to improve fitness scores.
The method further includes allowing users to select one or more of the generated solutions, viewing each solution alongside its respective fitness score, and inputting design style preferences, upon which the system generates multiple realistic views via a graphical user interface.
The method further includes transforming selected floor plans into 3D massing diagrams by employing a transformation processing unit.
The method further includes producing realistic images from the 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance the realism of generated images using a controlling unit.
The method further includes exporting a final 2D floor plan through a display unit.

The object of the present disclosure is to elevate architects' design workflow by improving both quality and efficiency.
Another object of the present disclosure is to empower architects to swiftly explore, optimize, and make informed decisions for space programming, planning, and form-finding. Achieve this through generating intelligent design alternatives aligned with specific goals and constraints.
Another object of the present disclosure is to reduce the time required to create conceptual-level floor plans. Simultaneously, enhance the overall quality of floor plans by mitigating human errors inherent in manual drafting processes.
Another object of the present disclosure is to establish a new pipeline that seamlessly integrates into the existing architectural workflow. Enable smooth export of designs to software platforms like SketchUp, ensuring fluid transition between different stages of the design process.
Yet another object of the present invention is to deliver an expeditious and cost-effective system for generating a floor plan.

To further clarify the advantages and features of the present disclosure, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which are illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail in the accompanying drawings.

BRIEF DESCRIPTION OF FIGURES

These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read concerning the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

Figure 1 illustrates a block diagram of a system for generating a floor plan, in accordance with an embodiment of the present disclosure;
Figure 2 illustrates a flow chart of a method for generating a floor plan;
Figure 3 illustrates a flow chart of the pipeline for generating a floor plan;
Figure 4 illustrates a block diagram for input data;
Figure 5 illustrates arepresentation of the room on the coordinate plane;
Figure 6 illustrates a crossover operation to create new solutions;
Figure 7 illustrates a translation along the Axes;
Figure 8 illustrates a rotation of room;
Figure 9 illustrates a dynamic resizing of rooms (Different Modification Steps);
Figure 10 illustrates a sample floor plan generation output;
Figure 11 illustrates multiple floorplans generated by the system;
Figure 12 illustrates a sample massing view from the generated floorplan;
Figure 13 illustrates a realistic house view generated by the system from the massing for a contemporary modern style home; and
Figure 14 illustrates a floor plan generated by the system is exported to Sketchup.

Further, skilled artisans will appreciate those elements in the drawings are illustrated for simplicity and may not have necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present disclosure. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein.

DETAILED DESCRIPTION:

To promote an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof.

Reference throughout this specification to “an aspect”, “another aspect” or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrase “in an embodiment”, “in another embodiment” and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.

The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises...a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.

Embodiments of the present disclosure will be described below in detail concerning the accompanying drawings.

Referring to Figure 1, a block diagram of a system for generating a floor planis illustrated in accordance with an embodiment of the present disclosure. The system 100 includes a data input unit (102) configured to receive architectural data and client preferences for room layouts and plot boundaries.

In an embodiment, a processing unit (104) is coupled to thedata input unit (102), theprocessing unit (104) configured to encapsulate the received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements, and initialize each room object by selecting a random value from an area range and ratio denominator range, and calculate dimensions based on the selected values.

In an embodiment, an evolutionary interface (106) is connected to theprocessing unit (104), theevolutionary interface (106) configured to generate sets of rooms in each iteration, each set representing potential room combinations as individual solutions. Then, evaluate the solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints. Thereafter, employ custom routines for weighted selection, crossover, and mutation of the solutions to improve fitness scores.

In an embodiment, a graphical user interface (108) is connected to theevolutionary interface (106), configured to allow users to select one or more of the generated solutions, view each solution alongside its respective fitness score, and input design style preferences, upon which the system generates multiple realistic views.

In an embodiment, a transformation processing unit (110)is connected to thegraphical user interface (108), configured to transform selected floor plans into 3D massing diagrams.

In an embodiment, a controlling unit (112)is in continuation with thetransformation processing unit (110), configured to produce realistic images from the 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance a realism of generated images.

In an embodiment, a display unit (114)is coupled to thecontrolling unit (112), configured to export a final 2D floor plan.

In another embodiment,the architectural data and client preferences are selected from a user-defined area of individual rooms, a ratio denominator to determine room proportions, and layout details of a boundary plot.

In another embodiment, the data input module incorporates a range specifier that allows users to input area ranges for rooms, define ratio denominator ranges, prevent a design of oblong rooms, and ensure compliance with preferred room dimension ranges, wherein the data input module further comprisesan extraction mechanism designed to obtain detailed specifications related to an area of dwelling.
In one embodiment, a ratio calculator is used to determine room proportions based on a given ratio denominator.
In one embodiment, a boundary detail retriever is used to collect information on a user-defined layout and plot boundaries.
In one embodiment, an adjacency mechanism is used to extract adjacency prerequisites for the rooms and utilize these prerequisites to influence a generation of optimal floor plans.

In another embodiment,theevolutionary interface (106), based on the encapsulated room objects, calculates room dimensions using length and width derived from the area and ratio denominator and determines initial room placements within given plot boundaries without exceeding the same.

In another embodiment, the fitness function quantifies a quality of each solution by assessing room boundary adherence, overlapping rooms, and adherence to room adjacency constraints, normalizing a score on a scale from 0 to 1, wherein the fitness function evaluates room boundary adherence by verifying whether room coordinates fall within user-defined plot boundaries, assigning a score of 1 for adherence and a score of 0 for any deviation, wherein the fitness function determines room overlap by computing an overlap area relative to total area of all rooms, wherein a score of 1 denotes no overlap and a score closer to 0 indicates higher overlap, wherein the fitness function calculates room adjacency based on given adjacency requirements, penalizing or rewarding solutions based on a distance between specified adjacent rooms and their adherence to adjacency constraints.

In another embodiment, the room object encapsulates data selected from room name, lower right diagonal coordinates, upper right diagonal coordinates, current area, current ratio denominator, area range, ratio denominator range, length, and width.

In one embodiment, a modification processing unit is coupled to theevolutionary interface (106), configured to iterate over various floor plan generations based on an Evolutionary technique paradigm and modify solutions in reference to the fitness score to evolve and produce improved solutions, wherein the modification processing unit comprisesa selection module configured to compute a sum of fitness scores of all chromosomes from a previous generation, determine a relative fitness of each chromosome based on its individual fitness and the computed sum, and perform weighted selection to determine parent chromosomes from a previous generation, prioritizing those with higher fitness scores while maintaining a possibility of selecting chromosomes with lower fitness scores to ensure solution diversity.
In one embodiment, a crossover module is configured to exchange room positions from parent chromosomes, creating offspring that inherit traits from a selected parent, and producing new solutions that seek to combine favorable attributes from parent solutions.
In one embodiment, a mutation module is configured to introduce variations in solutions through techniques including translation along axes, orthogonal rotation of rooms, and dynamic resizing of rooms.
In one embodiment, a visualization and checkpointing unit is coupled with the modification processing unit, configured to maintain a checkpoint list capturing best-scoring individuals/solutions during iterations, visualize an optimal floor plan by interpreting coordinates of boundary plots and rooms, and display the optimal floor plan to a user with appropriate annotations.
In one embodiment, a massing generation module is coupled with the visualization and checkpointing unit, configured to convert the optimal floor plan into an architectural massing by using room positions and elevation values, provides a 3D massing representation of a floor plan that can be manipulated in terms of rotation, scale, and zoom, and provide an exterior line view of the building as derived from the floor plan.
In one embodiment, a realistic house view generation module is configured to use conditional control to guide image generation diffusion models, generate realistic images of houses based on user-defined design style prompts and a 3D massing as a conditional image, and provide images that align with a range of design styles based on user preferences.
In one embodiment, an export module is coupled to the visualization and checkpointing unit, configured to convert the generated optimal floor plan into a format compatible with architectural modeling tools, facilitate export of the floor plan to modeling platforms, and allow for further refinement and adjustments in a modeling platform, bridging a gap between automated design and manual architectural refinement.

In another embodiment,the mutation module for translation along axes is configured to displace rooms along specific axes based on calculated delta movements, and update room coordinates through a variety of translation scenarios, such as lower bound coordinate modification, upper bound coordinate modification, or equal bound coordinate modification, wherein the mutation module for orthogonal rotation of rooms swaps length and width dimensions and their associated coordinates, effectively causing a 90-degree rotation, wherein the mutation module for dynamic resizing of rooms encompasses adjusting dimensions while keeping an area constant, modifying an area while retaining a ratio denominator constant and recalculating and adjusting both area and ratio denominator values simultaneously, wherein the dynamic resizing employs pseudorandom distributions to select new ratio denominators and area values, ensuring changes adhere to input constraints, and wherein the mutation module utilizes an adaptive mutation trigger to control a frequency of mutation, with a higher mutation chance during initial iterations and a decreasing probability in subsequent iterations, such that as solutions become more refined, the system limits perturbations to maintain optimal solutions, and wherein the adaptive mutation trigger is governed by parameters including a current generation iteration index, a total number of generations, and defined upper and lower bounds for mutation probability, thereby determining a final mutation probability for each iteration.

In another embodiment,the massing generation module calculates coordinates for all vertices of a room based on its elevation and boundaries to produce a three-dimensional representation, wherein the realistic house view generation module utilizes existing knowledge paired with a trainable copy, connected via zero-convolutions, to generate accurate and detailed house images based on a provided massing, and wherein the export module operates in a realm of Intelligence Amplification, assisting architects with pre-generated solutions and providing flexibility for further modifications.
Figure 2 illustrates a flow chart of a method for generating a floor plan. At step 202, method 200 includes receiving architectural data and client preferences for room layouts and plot boundaries through a data input unit (102).
At step 204, method 200 includes processing received architectural data and client preferences using a processing unit (104) by encapsulating said received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements, and initializing each room object by selecting a random value from an area range and ratio denominator range, and calculating dimensions based on said selected values.
At step 206, method 200 includes generating sets of rooms in each iteration, each set representing potential room combinations as individual solutions using an evolutionary interface (106) and evaluating said solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints thereby employing custom routines for weighted selection, crossover, and mutation of said solutions to improve fitness scores.
At step 208, method 200 includes allowing users to select one or more of the generated solutions, viewing each solution alongside its respective fitness score, and inputting design style preferences, upon which said system generates multiple realistic views via a graphical user interface (108).
At step 210, method 200 includes transforming selected floor plans into 3D massing diagrams by employing a transformation processing unit (110).
At step 212, method 200 includes producing realistic images from said 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance the realism of generated images using a controlling unit (112).
At step 214, method 200 includes exporting a final 2D floor plan through a display unit (114).

Figure 3 illustrates a flow chart of the pipeline for generating a floor plan.The proposed invention presents a comprehensive system aimed at transforming the conventional technical pipeline used by architects. This innovation seamlessly integrates user input concerning room layouts and plot boundaries, thereby generating intricate floor plans, massing diagrams, and lifelike visualizations. This inventive approach holds significant potential for streamlining building design processes and facilitating scalability across various operations.
The initial phase of the pipeline commences by gathering architectural data and client preferences for both the site and architectural program, as provided by the user. This data is encapsulated within a Room object, which serves as the foundational element for subsequent steps within the evolutionary technique. The representation of each room rectangle encompasses its left and right extreme coordinates. In each iteration, a set of rooms is generated, with all possible room combinations being considered as individual solutions. The generation process, depending on the defined population size, yields a specific number of solutions per generation. Throughout successive iterations, the aim is to progressively converge towards an optimal solution.
The fitness function assesses each solution, determining the inherent quality of the generated floor plan. The fitness score quantitatively evaluates several aspects: (1) adherence of rooms to buildable area boundaries, (2) avoidance of room overlaps, and (3) compliance with adjacency constraints.
Considering the current fitness score and the pursuit of increasing it, custom routines involving weighted selection and crossover are employed. Subsequently, solutions undergo mutations, using actions such as (1) vertical, horizontal, and diagonal translation, (2) room rotation, and (3) dynamic resizing of rooms.
The user gains the ability to view diverse solutions, each accompanied by its respective fitness score. Upon selecting a specific solution, the corresponding floor plan (i.e., the chosen solution) serves as the basis for creating a projected 3D massing. This 3D massing is used for generating realistic images through diffusion models for image generation. The user can input design style preferences as prompts, thereby generating a multitude of realistic views. The final 2D floor plan can even be exported to SketchUp, successfully integrating the pipeline into the existing workflow.
Figure 4 illustrates a block diagram for input data. Module 1 works around molding and extracting data for building their dream home. This data would prove vital in our successive steps to create floor plans, massing, and the overall view of their dream home. Details would be incisive on the area, ratio denominator (1/ratio denominator = length/width) for the rooms, and layout/boundary plot details. This information is entrenched firmly in our system so as to build everything from this constraint.
For the rooms, the adjacency requirements are also taken into account to provide valuable input while generating the optimal floorplans. Upon discretion, area ranges and ratio denominator ranges are also provided to avoid the possibility of oblong rooms and follow the structure dimension range for various rooms.
For the boundary/plot, the dimensions of the length and width are taken so as to create floor plans that would bind within the said constraints of the boundary.
Figure 5 illustrates a representation of the room on the coordinate plane. Module 2, focuses on levying the power of Computational Intelligence to account for and delineate various parameters considered to generate an optimal Floorplan.
The system works on a unique take of the Evolutionary Technique with custom routines to tackle the problem of evolving/generating an optimal solution for the use case. Adapting to the AI-based technique, the floorplan module includes many custom routines to handle different constraints to churn out an optimal solution.
With the aid of the Evolutionary Technique, the set of multiple solutions for a task is referred to as a "population" and each solution within the population is a "chromosome". The chromosome consists of the solution to the problem one is trying to solve (with various parameters). Given a numerical value of "population size", the population list would contain many amounts of chromosomes/possible solutions.
Initialization
In accordance with Module 1, the input data constraints for each room are collected and encapsulated into a Room object (Ratio Denominator Range, Area Range, Name, Adjacency List). This object forms the backbone of our successive steps.
Room initialization works as follows, a random value is picked from the Area Range and Ratio Denominator Range (part of the Input Constraint of the Room) and this value is assigned as the Room’s Current Area and Ratio Denominator Values respectively. Once the values of the Area and Ratio Denominator are set, dimensions (length, width) are calculated as follows (Equations 1 and 2, detail the calculation of the length and width from the Area and Ratio Denominator),

length = v( area×1/(ratio denominator))? (Equation 1)

width = area/length? (Equation 2)
The Dimensions of the Rooms form the backbone of how the Length, Width, Ratio Denominator, and Area are represented and furthermore accessed, optimized for the generation of optimal floorplans. Fig 5, shows the representation of the Room on the coordinate plane. Hence, for representation, only the bottom left extreme and the right extreme coordinates of the diagonal as the following can be inferred from the same,
length = x_2 - x_1? (Equation 3)
width = y_2 - y_1? (Equation 4)
After obtaining the dimensions, the coordinates for the initial room placement are selected by randomly picking out values (with the scale of length and width) and ensuring that it doesn't extend beyond the plot/layout boundary. First, the lower right diagonal coordinates are picked (random value within the limits of 0 and the difference between the boundary length/width and length/width of the room) and then the subsequent length/width is added to it to obtain the upper diagonal coordinates. This is done as follows,
x_1=random(0,boundarylength-roomlength)? (Equation 5)
y_1=random(0,boundarywidth - roomwidth)? (Equation 6)
x_2=x_1 + roomlength? (Equation 7)
y_2= y_1+roomwidth? (Equation 8)
Equations 5 and 6, work around selecting the lower right diagonal coordinates, and the upper right diagonal coordinates are calculated in Equations 7 and 8.
This process is repeated for all the rooms, thus the result is a chromosome (a possible solution for the task - here, accounts for the arrangement, position, and size of the rooms). Depending on the population size, the required number of chromosomes is constructed and added to the population list, which serves as a didactic initiation point for the generation of optimal floorplans. Each chromosome is represented by the following format:
chromosome/solution = (r_1,r_2,r_3,..,r_(total number of rooms)) ? (Equation 9)
where 'r' represents a room object
All the data specific to the Room is encapsulated in an object of the Class Room, this ensures the presence of granular information and can be accessed by the successive modules depending on their specific functions. The Room Class is as follows:The Room Class is as follows:
class Room:
Room Name
Lower Right Diagonal Coordinates (x1, y1)
Upper Right Diagonal Coordinate (x2, y2)
Current Area of the Room
Current Ratio Denominator of the Room
Area Range (lower and upper bound - part of Input Constraint)
Ratio Denominator Range (lower and upper bound - part of Input Constraint)
Length of the Room
Width of the Room

Fitness (Quantifying the score for each solution)

With every iterative step of the intelligent generation, the technique staunches upon a quantification step to criticize the overall progress of the generation. Quantifying the quality of each solution, the system is presented with a score, referred to here as the “fitness score”. If the score is high in magnitude, then it poses for a floor plan that is more optimal/feasible, if the score is low, then the floorplan generated through the solution is inferior. The scoring of the various floor plans presented by the chromosome will pave the way for the overall progress of how the technique can make intelligent updates to the existing solution to tune it in an optimal way to satisfy all the required constraints for the best possible floorplan.

Since building a floor plan is a meticulous process, the exhaustive constraints follow to determine the score and the progress of the technique are as follows:
The Rooms must stay within the boundary of the given layout
No Overlap
Adjacency Constraints of Various Rooms [Cohesion/Tightness of the Rooms]

The system's fitness score encompasses all of these constraints to instruct the technique to make necessary strides in achieving them. The fitness score is normalized to a scale of 0 - 1 (0 indicating the lowest/worst possible score and 1 indicating the best possible solution adhering to all the constraints)

The Rooms must stay within the boundary of the given layout

This condition is checked by verifying whether the coordinates of the rooms are within the boundary/plot coordinates given initially as a part of the input constraints. The following check reveals an output of binary value. If the room extends beyond the boundary, the boundary score is 0, else the boundary score is 1. All the rooms are checked in this fashion to verify whether the rooms adhere to the boundary constraint.
No Overlap
While generating floor plans, it's essential to avoid the potential case of overlaps. To address the same, an overlap score is computed which holds information regarding the overlap area. The overlap area is then divided by the total area of all the rooms, this signifies the relative ratio to which the overlap area compares to the overall room areas. The final score is the difference between 1 (no overlap) and the ratio. Hence, a score of 1 would yield no overlap, and a score of 0, would yield maximum overlap, For room_(1 ) and room_2, the overlap area is computed as follows,

?overlapx?_1=max(x_(1 ) ofroom_1,x_1 ofroom_2)? (Equation 10)
overlap x_2=min(x_2 of room_1,x_2 of room_2)? (Equation 11)
overlap y_1=max(y_1 of room_1,y_1 of room_2)? (Equation 12)
overlap? y?_2=min(y_2 of room_1,y_2 of room_2)? (Equation 13)
overlap area = (overlap x_2-overlap x_1)×(overlap y_2-overlap y_1)? (Equation 14)
As detailed form Equations 10 to 14, the overlap area is computed for all possible room combinations and then summed together to accumulate to the total overlap area present amongst the rooms, as described in Equation 15.
total overlap area = ?_i^(total rooms) ?_(j=i+1)^(total rooms) overlap area(room_i,room_j)? (Equation 15)

The final overlap score is shown in Equation 16,
overlap score = 1-(total overlap area)/(total area of the rooms)? (Equation 16)

Adjacency Constraints of Various Rooms
Given adjacency requirements, the system needs to penalize and reward that solution that adheres to the input adjacencies between the rooms. The encapsulated Room object carries information regarding the adjacency of the rooms. For those specified adjacent rooms, the midpoint distance between them is calculated and thereby incorporated into fitness to minimize the distance between them. Adjacency is calculated only if there isn’t an overlap between the rooms. For room_(1 ) and room_2, the adjacency is computed as follows,
midpoint x of room_1=x_1 of room_1+(length of room_1)/2? (Equation 17)
midpoint x of room_2=x_1 of room_2+(length of room_2)/2? (Equation 18)
midpoint y of room_1=y_1 of room_1+(width of room_1)/2? (Equation 19)
midpoint y of room_2=y_1 of room_2+(width of room_2)/2? (Equation 20)
adjacency distance = euclidean distance between midpoints of (room_1,room_2)
adjacency score =1 -(?_i^(total rooms) ?_(j=i+1)^(total rooms) adjacency distance(room_i,room_j))/maxscore
? (Equation 21)
All these constraints are tackled and quantified in the form of the fitness score, which discerns whether the generated plan is deemed optimal for our use case. The following functions culminate together to yield an accumulative score denoting all these various nuances that have to be scored upon. Quantitatively, a higher fitness score denotes the more optimal floor plan and a lower one denotes a suboptimal one (The values range between 0 and 1)

fitness score = boundary score × f(overlap score,adjacency score) ? (Equation 22)
Modification through Generation
The latent ability to pursue an Evolutionary Technique type paradigm is its ability to reflect on the changes and have a chance at improving in the coming iterations of the various floor plan generations/solutions. From a high-level view, the system adapts and modifies the solution with reference to the fitness score and evolves (generates solutions with higher fitness scores) in iterations to yield those solutions that would be closer to an optimal floor plan.
Selection of the best solutions
Depending on the population size, each generation is greeted with a population size number of solutions. With every iteration/generation, the goal is to subsequently move closer to finding an optimal solution. One way to stride forward in that direction is employing ways to capture those solutions/chromosomes with relatively higher fitness, all the while maintaining the diversity of solutions as well.
To initiate the generation of the new set of solutions, the more suitable parents/solutions from the previous solutions are extracted. They are extracted as follows:
Pseudocode:
Total Fitness Score ? Find the sum of the fitness of all the chromosomes in the previous generation
Weighted Fitness Ratio Array ? Empty Array
for chromosome in (previous generation):
Fitness of chromosome ? find the fitness of the chromosome
Relative Fitness of Chromosome ? (Fitness of the chromosome)/(Total Fitness Score)
Add to the Weighted Fitness Ratio Array ? Relative Fitness of Chromosome
Select 2 Parent Chromosome based on random Weighted Selection of the Weighted Fitness Ratio Array
The advantage of this method type selection is that the parents are chosen in a way that maximum weight/probability is given to the parents with the highest fitness score, but it also subsumes diversity in solutions by ensuring that the lower fitness score chromosomes also have a way to be chosen as a parent for the future generations. By maintaining diversity and also the most probable fitness parent, the system continues its search for the optimal floor plan.
Figure 6 illustrates a crossover operation to create new solutions. The Crossover technique is used to exchange information from the parents' side and pass it on to the next generation of chromosomes. Since the selection of the parents would most probably employ the best-fit individuals, a crossover of those solutions would lead to new chromosomes/offspring that would retain certain portions of the parent which worked right (since the parents are the fit individuals in the previous generation). The crossover technique used in the system exchanges the room positions from the two parents and hence, creates a resultant new chromosome/solution/offspring, thereby capturing the best traits from the parent solutions and creating a new set of solutions that would work together to achieve an optimal generation result.
Variation/Mutation of Solutions
In every generation, to reach the optimal floor plan generation goal, there are operations performed on each chromosome that would help in progressing the overall strides to achieve a higher fitness score.For generating optimal floorplans, mutation functions form a fundamental and ardent building block that would guide the chromosome variations. Applying functions on the given chromosome/solution would improve its ability to pace toward a relatively higher fitness score.
Factors considered for progressive strides during variations functions are:
- Design Constraints of Area and Dimensions
- Orientation of Rooms
- Rotation of Rooms
These are the following functions of Mutations/Variations performed on the Chromosomes:

Figure 7 illustrates a translation along the Axes.
Translation along the Axes
? For each room, they are translated/moved along the axes to change the layout and subsequently alter the overall representation of the resultant floor plan
? To account for the movement, there are 3 possible scenarios by which the rooms can be translated: Along the X-Axis, Along the Y-Axis, and movement through both Axes (Diagonal-like Displacement)
? For the axis 'i' of room 'R', the heuristic translation is as follows:
a. Compute the minimum limit <- Distance from room_lower_i coordinate (lower coordinate of the room along the axis i) to the lower bound coordinate of the plot boundary, say 'b_lower_i'
b. Compute the maximum limit <- Distance from the room_upper_i (upper coordinate of the room along the axis i) to the upper bound coordinate of the plot boundary, say 'b_upper_i'
c. After obtaining both extremities, a uniform probabilistic distribution is maintained for the given limits and the delta movement values are extracted from this. This value signified the change in position from the previous reference point
d. The extracted common value is then added to the coordinates of the room along the specified 'i' axis and thereby displaced to a new position
minimum limit=|? i?_1 of room-i_1 of boundary |? (Equation 23)
maximum limit=|? i?_2 of room-i_2 of boundary |? (Equation 24)
delta movement along axis_i=random(uniform(-minimum limit,maximum limit)
? (Equation 25)
new i_1 of room=old i_1+delta movement along axis_i? (Equation 26)
new i_2 of room=old i_2+delta movement along axis_i? (Equation 27)
? The Equations from 23 to 27 illustrate how the coordinates are updated for any axis ‘i’ (x/y axis) to signify the translation/movement of the rooms within the layout
? Separate functions are executed for the translating along the x-axis, y-axis, and both. This ensures diversity in the solution and engenders a whole new possibility of obtaining and reverencing along the progress of an optimal floorplan score

Figure 8 illustrates a rotation of room.
Rotation of Rooms
? This function presents a dynamic attribute concerning the position and orientation of the rooms
? The function embellishes by rotating the rooms orthogonally to meet/form a concise floorplan capable of satisfying the various constraints
? For rotating the rooms, the length and width of the rooms are swapped (including their coordinates), which would result in an orthogonal rotation
? For a Room ‘r’, the following would be changed on rotation (Equations 28 to 33),
new x_1=old y_1? (Equation 28)
new x_2=old y_2? (Equation 29)
new y_1=old x_1? (Equation 30)
new y_2=old x_2? (Equation 31)
new length=old width? (Equation 32)
new width=old length? (Equation 33)
? By adding capabilities of rotation, the technique is aided with a diverse set of functions to meet and treat solutions that would be suitable for an optimal floor plan

Figure 9 illustrates a dynamic resizing of rooms (Different Modification Steps).
Dynamic Resizing of Rooms
Another novel routine that would suit in favor of the overall goal of optimal floor plan generation is the addition of the Dynamic Resizing Module. The granularity of the system remains in its ability to encapsulate Room Specific information in its class and the functions that would serve to diversify and vary solutions to create novel upbringings of floorplans. This function invigorates the ability of the system to change the current characteristics of the rooms while also maintaining the input constraints as the source of truth. The functions tackle three scenarios, namely, "Change Dimensions (length and width) for a constant Area", "Change Area for a constant Ratio Denominator Value", and "Recalculating and adjusting to new Area and Ratio Denominator Values" in the following modules
Change Dimensions for a Constant Area:
This module works around focusing on dynamically adjusting the dimensions based on a constant Area. Hence, this module would adjust the dimensions (with an account of the Ratio Denominator) in a way constant Area is maintained by the Room
With the Room object encapsulating the Room-specific information, a new Ratio Denominator is randomly picked from a Uniform Distribution of the range (the user provides lower and upper bound as a part of the input) provided by the input constraints
If r_d lower and r_d upperare the Input Ratio Denominator lower bound and upper bound provided by the Input Constraints respectively, then,

Ratio Denominator Current_new=random(uniform(r_d lower,r_d upper)))? (Equation 34)
For the existing Area and the new Ratio Denominator of the room, the dimensions are recalculated as in Equations 1 and 2
Change Area for a Constant Ratio Denominator:
This module works around maintaining the constant Ratio Denominator, by adjusting the dimensions for a newly found Area value. This module acts to focus on the dynamicity of the system
With the Room object encapsulating the Room-specific information, a new Area is randomly picked from a Uniform Distribution of the range (the user provides lower and upper bound as a part of the input) provided by the input constraints
If A_lower and A_upperare the Input Area lower bound and upper bound provided by the Input Constraints respectively, then,
Area Current_new=random(uniform(A_lower,A_upper)))? (Equation 35)
? For the existing Area and the new Ratio Denominator of the room, the dimensions are recalculated as in Equations 1 and 2
? Recalculating and adjusting to new Area and Ratio Denominator Values
? The module combines the dynamicity of both the above modules by changing both the Area and Ratio Denominator about the Room
? By obtaining the new Ratio Denominator and Area value by randomly picking from the Uniform Distribution of the ranges (the user provides lower and upper bound as a part of the input), the module strives to work around dynamically modifying both the Area and Ratio Denominator
? Picking a new Ratio Denominator and Area Value is the combination of Equations 34 and 35
? For the existing Area and the new Ratio Denominator of the room, the dimensions are recalculated as in Equations 1 and 2
On obtaining the newly formulated length and width (for all the above 3 modules), the difference between the old dimensions and the new dimensions is calculated (the difference is referred to as the 'delta length' and 'delta width’).
delta_length=length_new-length_old? (Equation 36)
delta_width=width_new-width_old? (Equation 37)
The process of updating the existing room dimensions (and coordinates) is split into 3 possibilities,
a. Lower Bound Coordinate Modification: In this way, the delta value is subtracted from the lower bound coordinate, belonging to the specific dimension i.e. for the width, 'delta width' is subtracted from the lower x coordinate value
?newx?_1=oldx_1-delta_length? (Equation 38)
new y_1=old y_1-delta_width? (Equation 39)
b. Upper Bound Coordinate Modification: In this way, the delta value is added to the upper bound coordinate, belonging to the specific dimension i.e. for the width, 'delta width' is added to the upper bound x coordinate value
?newx?_2=oldx_2+delta_length? (Equation 40)
new y_2=old y_2+delta_width? (Equation 41)
c. Equal Bound Coordinate Modification: In this way, the equal half of the delta value is added to both the upper and lower bound coordinate, belonging to the specific dimension i.e. for the width, half of 'delta width' is added to the upper bound x coordinate value and another half to the lower bound x coordinate value
?newx?_1=oldx_1-(delta_length)/2? (Equation 42)
new y_1=old y_1-(delta_width)/2? (Equation 43)
?newx?_2=oldx_2+(delta_length)/2? (Equation 44)
new y_2=old y_2+(delta_width)/2? (Equation 45)
he modules mentioned above, "Change Dimensions (length and width) for a constant Area", "Change Area for a constant Ratio Denominator Value", and "Recalculating and adjusting to new Area and Ratio Denominator Values", are selected depending on a probabilistic based randomized manner to encourage different modules to work on diversifying and creating a broader search space for the optimal floor plan.
Hence, these modules add dynamicity to the system and provide it a landscape to tune the particular system to yield a higher score, which eventually leads to optimal floor plans.
The selection of which mutation function to use on the chromosome/solution is guided with the help of a heuristic weighted selection module. With the aid of the weighted selection, priorities are maintained for certain functions over others, hence supplanting the system and users with the possibility of prioritizing certain functions over others. Though it is to be noted, even with weighted selection, there still exists the possibility of choosing the lower preferred modules because of the random selection (based on weight priorities), hence allowing the system to add that diversifying authenticity to its solution set (as described in equation 46, where w1, w2, w3 are the weights and f_1, f_2, f_3are the mutation functions)
mutation function = random(w1×f_1,w2×f_2,w3×f_3)? (Equation 46)
To trigger the mutation execution (mutation is not always triggered, supplementing the randomness factor that can be employed to create diverse solutions), an Adaptive Mutation Trigger function is set in place to generate the Mutation Probability. The functionality of the Adaptive Mutation Trigger is to have a high chance of mutation in the initial iterations and as multiple iterations progress, reduce the frequency of mutations (this is because as generations/iterations loop by, the solutions would be all the more refined, matured and tuned to achieve higher fitnesses and optimal floor plans), so as to not perturb the more refined and matured solutions. The Adaptive Mutation is contingent on Equation 47. A further refinement of setting the bounds of the probability is done so as to keep the overall mutation probability value in check and is limited between the two hyperparameters, "Maximum Mutation Probability" and "Minimum Mutation Probability", as described in Equation 48. This yields a final mutation probability, if the generated random value is within the Mutation probability, the Mutation is triggered.
m_a=m_p×(1-g_idx/g_t )? (Equation 47)
m_final=max(m_min,min(m_max,m_a))? (Equation 48)
where,
m_a? AdaptiveMutationProbability,
g_idx? CurrentGenerationIterationIndex,
g_t? TotalNumberofGenerations (always=1),
m_min? MinimumMutationProbability,
m_max? MaximumMutationProbability,
m_final? FinalMutationProbability
The Variation/Mutation of Solutions helps in diversifying the solutions and perturbing the rooms in various routines to create solutions that would create optimal floorplans.
Visualization and Checkpointing
During the iteration of the system, a checkpoint list is maintained to capture the best-scoring individuals/solutions. This list maintains the best-fit solutions through which one can see the growth and select the required floor plan. After choosing the best solution (highest fitness score), given the coordinate of the boundary plot and all the rooms (encapsulated in the Room Class), the following result is then visualized (the rooms are also annotated) to present to the user with the optimal floor plan.

Figure 10 illustrates a sample floor plan generation output.
Figure 11 illustrates multiple floorplans generated by the system.
Figure 12 illustrates a sample massing view from the generated floorplan. This module revolves around generating a base Architectural Massing from the generated optimal floorplan. The generated floor plan contains a set of coordinate values for the rooms, they are positioned in a place accordingly. After incisive position, depending on the elevation of each room (coordinate along the z-axis), they are subsequently projected up iteratively. The result is the Massing of the generated optimal floorplan, the Massing can be rotated, scaled, or zoomed as required to view the different perspectives involved in the same. The culmination of the module is the exterior line view of the projected building from the floorplan. Hence, this module works on pedantic aspects to relive and revitalize the floorplan to create a projected 3D Massing which would be tantamount to a base representation of the designed building.
For a Room ‘r’ with elevation ‘e’, the following are the coordinates of all its vertices,
(x_1,y_1,0),(x_2,y_1,0),(x_2,y_2,0),(x_1,y_2,0),(x_1,y_1,e),(x_2,y_1,e),(x_2,y_2,e),(x_1,y_2,e)
Figure 13 illustrates a realistic house view generated by the system from the massing for a contemporary modern style home. This module works on Generating Realistic House Views from the Massing obtained from the previous module. This module leverages the prowess of Conditional Control to Image Generation Diffusion Models - (L. Zhang and M. Agrawala, “Adding conditional control to text-to-image diffusion models,” arXiv [cs.CV], 2023). As mentioned by the authors (L. Zhang and M. Agrawala, “Adding conditional control to text-to-image diffusion models,” arXiv [cs.CV], 2023), the architecture reconciles the addition of Conditional reference, which guides the existing Image Generation modules to work from and tune further based on the input prompt. With the aid of a Trainable Copy and a Locked Copy, the preexisting knowledge seems to remain unperturbed and the Trainable Copy is connected by a bridge of Zero-Convolutions. Hence, the conditional pairs of Images would provide a realistic Image Generation contingent on the conditional reference point. In our system, the base Image generation has been tuned using multiple Home Design Style Prompts and Image pairs (using, R. Gal et al., “An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion,” arXiv [cs.CV], 2022), the Massing generated in the previous module engenders the Conditional Image, and the result from the Model is a Realistic House based on the Design Style preferred by the user. The Realistic Image Generated from the Massing strikes an uncanny resemblance to various design styles and creates a completely new perspective on viewing their dream home. The user can provide various design styles to their liking to generate a plethora of Realistic Views.
This end-to-end system of accepting user input for rooms and plot boundaries to generate Floor Plans, Massing, and Realistic House views forms a novel take in the field of Architecture. This pipeline would help streamline the process of building one's home and would help many entities to scale for many more processes. This system starts from the ground up to generate an optimal Floor Plan based on the constraints of 3D Massing and eventually a Realistic View of their dream home goes to unprecedented stretches of creating stunning and creative designs, and generations.
Figure 14 illustrates a floor plan generated by the system is exported to Sketchup. This module invigorates from the extended functionality point of view. This module works around exporting the generated optimal floor plan to an Architectural Modelling tool, called Sketchup. Hence, providing an option for Architects to have the ability to make finer adjustments if required. This scope adds the system within the realm of Intelligence Amplification and aids the Architects by generating solutions for various plans and offering an extended functionality to tune/make finer adjustments when required. Integrating the option to add export to Sketchup just makes our system all the more versatile and accommodative to the entire Design Pipeline.
The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.

Benefits, other advantages, and solutions to problems have been described above about specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.

, Claims:1. A system for generating a floor plan, the system comprises:

a. a data input unit configured to receive architectural data and client preferences for room layouts and plot boundaries;
b.a processing unit coupled to said data input unit, said processing unit configured toencapsulate said received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements, and initialize each room object by selecting a random value from an area range and ratio denominator range, and calculate dimensions based on said selected values;
c.an evolutionary interface connected to said processing unit, said evolutionary interface configured to:
i) generate sets of rooms in each iteration, each set representing potential room combinations as individual solutions;
ii) evaluate said solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints;
iii) employ custom routines for weighted selection, crossover, and mutation of said solutions to improve fitness scores;
d. a graphical user interface connected to said evolutionary interface, configured to allow users to select one or more of the generated solutions, view each solution alongside its respective fitness score, and input design style preferences, upon which said system generates multiple realistic views;
e. a transformation processing unit connected to said graphical user interface, configured to transform selected floor plans into 3D massing diagrams;
f. a controlling unit in continuation with said transformation processing unit, configured to produce realistic images from said 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance a realism of generated images; and
g.a display unit coupled to said controlling unit, configured to export a final 2D floor plan.

2. The system as claimed in claim 1, wherein said architectural data and client preferences are selected from a user-defined area of individual rooms, a ratio denominator to determine room proportions, and layout details of a boundary plot.

3. The system as claimed in claim 1, wherein said data input module incorporates a range specifier that allows users to input area ranges for rooms, define ratio denominator ranges, prevent a design of oblong rooms, and ensure compliance with preferred room dimension ranges, wherein said data input module further comprises:
i) an extraction mechanism designed to obtain detailed specifications related to an area of dwelling;
ii) a ratio calculator to determine room proportions based on a given ratio denominator;
iii) a boundary detail retriever to collect information on a user-defined layout and plot boundaries; and
iv) an adjacency mechanism to extract adjacency prerequisites for said rooms and utilize these prerequisites to influence a generation of optimal floor plans.

4. The system as claimed in claim 1, wherein said evolutionary interface, based on said encapsulated room objects, calculates room dimensions using length and width derived from said area and ratio denominator and determines initial room placements within given plot boundaries without exceeding the same.

5. The system as claimed in claim 1, wherein said fitness function quantifies a quality of each solution by assessing room boundary adherence, overlapping rooms, and adherence to room adjacency constraints, normalizing a score on a scale from 0 to 1;
wherein said fitness function evaluates room boundary adherence by verifying whether room coordinates fall within user-defined plot boundaries, assigning a score of 1 for adherence and a score of 0 for any deviation;
wherein said fitness function determines room overlap by computing an overlap area relative to total area of all rooms, wherein a score of 1 denotes no overlap and a score closer to 0 indicates higher overlap; and
whereinsaid fitness function calculates room adjacency based on given adjacency requirements, penalizing or rewarding solutions based on a distance between specified adjacent rooms and their adherence to adjacency constraints.

6. The system as claimed in claim 1, wherein said room object encapsulates data selected from room name, lower right diagonal coordinates, upper right diagonal coordinates, current area, current ratio denominator, area range, ratio denominator range, length, and width.

7. The system as claimed in claim 1, further comprises:
a. a modification processing unit, coupled to said evolutionary interface, configured to iterate over various floor plan generations based on an Evolutionary technique paradigm and modify solutions in reference to said fitness score to evolve and produce improved solutions, wherein said modification processing unit comprises:
i) a selection module configured to compute a sum of fitness scores of all chromosomes from a previous generation, determine a relative fitness of each chromosome based on its individual fitness and said computed sum, and perform weighted selection to determine parent chromosomes from a previous generation, prioritizing those with higher fitness scores while maintaining a possibility of selecting chromosomes with lower fitness scores to ensure solution diversity;
ii)a crossover module configured to exchange room positions from parent chromosomes, creating offspring that inherit traits from a selected parents, and producing new solutions that seek to combine favorable attributes from parent solutions; and
iii) a mutation module configured to introduce variations in solutions through techniques including translation along axes, orthogonal rotation of rooms, and dynamic resizing of rooms;
b. a visualization and checkpointing unit, coupled withsaid modification processing unit, configured to maintain a checkpoint list capturing best-scoring individuals/solutions during iterations, visualize an optimal floor plan by interpreting coordinates of boundary plots and rooms, and display said optimal floor plan to a user with appropriate annotations;
c. a massing generation module, coupled with said visualization and checkpointing unit, configured to convert said optimal floor plan into an architectural massing by using room positions and elevation values, provides a 3D massing representation of a floor plan that can be manipulated in terms of rotation, scale, and zoom, and provide an exterior line view of the building as derived from said floor plan;
d. a realistic house view generation module, configured to use conditional control to guide image generation diffusion models, generate realistic images of houses based on user-defined design style prompts and a 3D massing as a conditional image, and provide images that align with a range of design styles based on user preferences; and
e. an export module, coupled to said visualization and checkpointing unit, configured to convert said generated optimal floor plan into a format compatible with architectural modeling tools, facilitate export of said floor plan to modeling platforms, and allow for further refinement and adjustments in a modeling platform, bridging a gap between automated design and manual architectural refinement.

8. The system as claimed in claim 7, wherein said mutation module for translation along axes is configured to displace rooms along specific axes based on calculated delta movements, and update room coordinates through a variety of translation scenarios, such as lower bound coordinate modification, upper bound coordinate modification, or equal bound coordinate modification;
whereinsaid mutation module for orthogonal rotation of rooms swaps length and width dimensions and their associated coordinates, effectively causing a 90-degree rotation;
wherein said mutation module for dynamic resizing of rooms encompasses adjusting dimensions while keeping an area constant, modifying an area while retaining a ratio denominator constant and recalculating and adjusting both area and ratio denominator values simultaneously, wherein said dynamic resizing employs pseudorandom distributions to select new ratio denominators and area values, ensuring changes adhere to input constraints; and
wherein said mutation module utilizes an adaptive mutation trigger to control a frequency of mutation, with a higher mutation chance during initial iterations and a decreasing probability in subsequent iterations, such that as solutions become more refined, said system limits perturbations to maintain optimal solutions, and wherein said adaptive mutation trigger is governed by parameters including a current generation iteration index, a total number of generations, and defined upper and lower bounds for mutation probability, thereby determining a final mutation probability for each iteration.

9. The system as claimed in claim 7, wherein said massing generation module calculates coordinates for all vertices of a room based on its elevation and boundaries to produce a three-dimensional representation;
wherein said realistic house view generation module utilizes existing knowledge paired with a trainable copy, connected via zero-convolutions, to generate accurate and detailed house images based on a provided massing; and
whereinsaid export module operates in a realm of Intelligence Amplification, assisting architects with pre-generated solutions and providing flexibility for further modifications.

10. A method for generating a floor plan, the method comprises:

a. receiving architectural data and client preferences for room layouts and plot boundaries through a data input unit;
b. processing received architectural data and client preferences using a processing unit byencapsulating said received architectural data and client preferences into room objects, each room object comprising a ratio denominator range, an area range, a name, and an adjacency list, forming foundational elements, and initializing each room object by selecting a random value from an area range and ratio denominator range, and calculating dimensions based on said selected values;
c. generating sets of rooms in each iteration, each set representing potential room combinations as individual solutions using an evolutionary interface and evaluating said solutions using a fitness function, which quantifies adherence to buildable area boundaries, avoidance of room overlaps, and compliance with adjacency constraints thereby employing custom routines for weighted selection, crossover, and mutation of said solutions to improve fitness scores;
d. allowing users to select one or more of the generated solutions, viewing each solution alongside its respective fitness score, and inputting design style preferences, upon which said system generates multiple realistic views via a graphical user interface;
e. transforming selected floor plans into 3D massing diagrams by employing a transformation processing unit;
f. producing realistic images from said 3D massing diagrams using diffusion models that leverage lighting, shadow, and texture parameters to enhance a realism of generated images using a controlling unit; and
g.exporting a final 2D floor plan through a display unit.