Description:Form 2

THE PATENTS ACT, 1970
COMPLETE SPECIFICATION
(See section 10 and rule 13)

Environment Meteorological Drone
for Comprehensive Environmental Data Collection and Analysis

APPLICANT NAME: INSTITUTE OF ENGINEERING & MANAGEMENT

ADDRESS: INSTITUTE OF ENGINEERING & MANAGEMENT, SALT
LAKE ELECTRONICS COMPLEX SECTOR – V, SALT
LAKE, KOLKATA, PIN – 700091.

The following specification particularly describes the invention
and
the manner in which it is to be performed

Field of the invention:
The present invention discloses unmanned aerial vehicles (UAVs) intended for complete environmental data gathering and analysis. This seeks to use a dynamic and adaptable UAV platform to completely reshape the field of environmental monitoring. More specifically, the drone is configured using a precision-tuned aerodynamic frame, integrated with the DG1 propulsion engine, custom-designed propellers for flight accuracy, and a secure assembly of key components such as the flight controller, GPS module, and an array of cutting-edge environmental sensors. The central to the flight control system is a specialized PID (Proportional-Integral-Derivative) control algorithm, executed by a microcontroller, which provides dynamic stability and precise maneuverability—even in harsh or wind-deflected environments. This enables smooth, controlled flight over long distances and under varying meteorological conditions. For forecasting and deep data analysis, the system includes an on-board computer that encodes sensor readings as vectors, processes them locally using lightweight AI models, and stores them for further downstream analysis.The system operates with accuracy and graph charting, it can send an infinite amount of data of any available environment.

Background of the Invention:
Traditional methods for collecting environmental data are often beset by fixed locations, parameter restrictions, and difficulties with real-time monitoring. These drawbacks are intended to be overcome by the All-Environment Meteorological Drone that provides an affordable, adaptable way to get beyond the limitations of traditional methods. The drone is made to meet a variety of requirements and is meant to serve as a new benchmark in the field of environmental data collection. With accuracy and graph graphing, it can send a limitless quantity of data in any available environment.
US20230003919A1- Weather Drone discloses a real-time analysis feature during data recording which made possible by an integrated processor and memory system. It does this by measuring weather parameters repeatedly using a first sensor. Drones are used in traditional meteorological data collection methods, but these methods have drawbacks such as incorrect readings, malfunctioning sensors, and the possibility of gathering erroneous data over time. The disclosed drone takes a proactive stance in response to these ongoing difficulties. During the recording phase, the CPU actively examines the data and goes so far as to stop storing more data in the memory if the recorded values are higher than predefined thresholds. The implementation of this rigorous real-time analysis and prevention technique aims to address the common problems related to the storage of potentially erroneous or damaged meteorological data.
Whereas the present invention surpasses US Patent Application US20230003919A1 in several key hardware aspects, making All-Environment Meteorological Drone more advanced and versatile. Unlike the US patent, which primarily focuses on a processor-controlled threshold-based data collection system with limited weather sensors (such as temperature and humidity), our drone integrates a broader range of cutting-edge sensors, including the MQ2 gas leakage sensor, MQ135 air quality sensor, and a raindrop sensor, allowing for comprehensive environmental analysis. Additionally, while the US patent relies on basic memory storage and a wireless communication module, our system enhances data accessibility with an LCD for real-time visualization, memory card integration, and an MQTT-based network for seamless transmission via Node-RED servers. Moreover, the present drone is designed for long-range stability with a precisely tuned frame, premium LiPo batteries for extended flight time, and robust wind deflection resistance—capabilities not explicitly covered in the US patent. These superior hardware design in our patent discloses a more advanced and practical solution for meteorological data collection in diverse environments.
US20200001735A1 discloses monitoring system, base station and control method of a drone presents a base station, a complete drone monitoring system, and a drone control method. The system aims to increase overall dependability, optimize charging efficiency, and adapt to a variety of situations by addressing issues related to automated monitoring equipment. The automated monitoring tools available today have a number of drawbacks. Problems include unstable drone landings, restricted charging capabilities, possible short circuits, subpar operation in harsh weather, and a lack of environmental adaptation. Furthermore, battery capacity limits how long monitoring equipment may operate. In order to address current issues, the revealed system integrates a drone with a base station and adds cutting-edge functions. Three major breakthroughs include a temperature control system that adapts to the environment, an exact landing mechanism with the help of a locating device, and optimized charging conditions based on battery specifications. Efficient charging is guaranteed by the base station's charging equipment, and the environmental monitoring system automatically modifies drone parameters in response to outside circumstances. When combined, these enhancements improve operating flexibility, charging efficiency, and dependability across a range of monitoring circumstances.
Whereas the present invention significantly outperforms US Patent Application US20200001735A1 in terms of hardware design and capabilities. While US20200001735A1 primarily focuses on drone monitoring and automated charging via a base station, our All-Environment Meteorological Drone is built with a superior sensor suite and an independent data processing system. Our drone integrates six specialized sensors, including the MQ2 gas leakage sensor, MQ135 air quality sensor, a raindrop sensor, and the DHT11 temperature and humidity sensor, providing a broader range of environmental data collection compared to the generic sensors mentioned in the US patent. Additionally, our design incorporates an LCD for real-time data visualization, whereas the US patent relies on remote transmission to a monitoring station without direct user interaction. Our drone is also equipped with an MQTT and Node-RED-based network for efficient wireless data transfer, making it more versatile and independent, unlike US20200001735A1, which depends on a fixed base station for communication and power management. Furthermore, our aerodynamically optimized frame and premium LiPo batteries ensure superior flight stability, longer endurance, and resistance to harsh weather conditions. With provisions for future upgrades such as AI-based autonomous piloting and more powerful microcontrollers, our patent presents a far more advanced and adaptable hardware system than US20200001735A1.
KR102580312B1 Observation data quality inspection apparatus and observation data quality inspection method using the same. The apparatus for checking the quality of observed data and the procedure for doing so are the subjects of the current invention. Obtaining observation data, carrying out a range check to remove mistake values, running an internal consistency test, inspecting the system step-by-step based on change thresholds, and utilizing previously stored model data for a spatial consistency check are the many stages involved in the system. For accurate weather forecasts and services, meteorological data quality is essential. The accuracy of data, however, can be impacted by intrinsic flaws in communication, observing behavior, and observational equipment. These mistakes make it difficult to guarantee that meteorological services exclusively employ high- quality data. The present invention suggests a thorough process for evaluating the caliber of data that has been observed. Step-by-step inspection based on change thresholds, an internal consistency test to verify adherence to physical relationships, a range check to eliminate values that depart from the sample group, and a spatial consistency check utilizing pre-stored model data are all part of it. By taking a multi-step strategy, the total quality of meteorological data will be improved, which will increase the dependability of weather-related services.

The present invention offers several key advantages over KR Patent KR102580312B1, particularly in terms of hardware integration, sensor capabilities, and real-time data processing. The Korean patent primarily focuses on a quality inspection system for meteorological data, utilizing computational methods to filter errors in climate observations. In contrast, our All-Environment Meteorological Drone is a fully autonomous aerial system designed to actively collect, process, and transmit environmental data in real time. Our drone is equipped with six specialized sensors, including the MQ2 gas leakage sensor, MQ135 air quality sensor, raindrop sensor, and DHT11 temperature and humidity sensor, which provide a broader range of environmental monitoring capabilities compared to the static ground-based sensors used in the KR patent. Additionally, our design incorporates an LCD for direct data visualization, ensuring real-time field usability, whereas the KR patent relies on external data processing units without an immediate feedback mechanism. Furthermore, our drone integrates an MQTT and Node-RED-based network for efficient wireless data transmission, making it more versatile and capable of operating in remote locations without relying on external infrastructure. The LiPo batteries of our drone ensure longer flight endurance, stability in harsh weather conditions, and superior adaptability compared to the KR patent's ground-based sensor network. Our system is also designed for future scalability, with provisions for AI-based autonomous piloting and machine learning-driven weather prediction, making it a more advanced and comprehensive hardware solution than KR102580312B1.
Patent: JP2020153792A: Meteorological observation system and flight control system for drone
The invention aims to improve the safety of unmanned aircraft, such as drones, during flight. It relates to a meteorological observation system and a drone flight control system. Using data from the drone's control system while in flight, the meteorological observation system uses a drone to fly nearly straight above or below in a vertical direction to monitor wind directions and speed. This technology can be used with a drone flight control system that chooses a second drone's flight path based on wind data collected by a first drone travelling a predefined path. Drone safety must be ensured in a variety of applications, such as alpine accident rescue scenarios. However, it has been difficult to get precise wind direction and speed data at the altitude of drone flight, and current drone control systems might not be able to avert accidents in the event of unforeseen weather. A drone is flown vertically as part of the planned meteorological observation system to measure the wind at a given altitude. These meteorological data are also used by the flight control system to choose safe flight paths for incoming drones. Safety is improved by the use of lightweight drones, emergency landing locations, and the ability to restrict flight range using a wire.
The present invention discloses over KR Patent KR102580312B1, particularly in terms of hardware integration, sensor capabilities, and real-time data processing. The Korean patent primarily focuses on a quality inspection system for meteorological data, utilizing computational methods to filter errors in climate observations. In contrast, our All-Environment Meteorological Drone is a fully autonomous aerial system designed to actively collect, process, and transmit environmental data in real time. Our drone is equipped with six specialized sensors, including the MQ2 gas leakage sensor, MQ135 air quality sensor, raindrop sensor, and DHT11 temperature and humidity sensor, which provide a broader range of environmental monitoring capabilities compared to the static ground-based sensors used in the KR patent. Additionally, our design incorporates an LCD for direct data visualization, ensuring real-time field usability, whereas the KR patent relies on external data processing units without an immediate feedback mechanism. Furthermore, our drone integrates an MQTT and Node-RED-based network for efficient wireless data transmission, making it more versatile and capable of operating in remote locations without relying on external infrastructure. The aerodynamic frame and premium LiPo batteries used in our drone ensure longer flight endurance, stability in harsh weather conditions, and superior adaptability compared to the KR patent's ground-based sensor network. Our system is also designed for future scalability, with provisions for AI-based
Therefore, to the best of our knowledge, none of the above-mentioned prior art attempts, individually or collectively propose the system and embodiments indicated and disclosed by the present invention

Object of the invention:
The primary objective of the present invention is to develop an advanced environmental drone system that seamlessly integrates hardware-level stability with intelligent data acquisition and processing capabilities.
Another objective is to enable the drone to operate stably in dynamic and rough environmental conditions using a microcontroller-based flight controller running a specialized PID control algorithm.
Another objective is to incorporate a wide array of environmental sensors (such as temperature, humidity, and air quality sensors), managed by a microcontroller, to ensure real-time, high-resolution data collection across varying terrains and climates.
A further objective is to integrate an onboard computer capable of encoding environmental data as vectors, performing lightweight inference using embedded models, and preparing data for secure transmission and downstream analysis.
Yet another objective of the invention is to provide automatic synchronization with a base station, where the collected data undergoes advanced processing through a hybrid CNN-LSTM architecture for predictive modeling, followed by in-depth analytics for pattern recognition and trend analysis.
Further the objective of the present invention is to establish a dual-mode communication architecture, comprising both an offline ad-hoc network (with a personalized broker and subscriber model) and an internet-enabled mode, thus ensuring uninterrupted data transmission and system control.
Further objective is to enable real-time visualization and full control of the drone’s data and operation through an interactive dashboard, facilitating field deployment, monitoring, and decision-making in critical environmental research and applications.

Summary of the Invention:
The present invention relates to an Environmental Meteorological Drone system designed for comprehensive environmental data collection, seamless processing, and advanced forecasting analysis across a wide range of weather conditions. The invention combines a finely engineered aerial platform with an intelligent software and server infrastructure. The drone is constructed using a precision-tuned aerodynamic frame, integrated with the DG1 propulsion engine, custom-designed propellers for flight accuracy, and a secure assembly of key components such as the flight controller, GPS module, and an array of cutting-edge environmental sensors. In an aspect of the invention , the central to the flight control system is a specialized PID (Proportional-Integral-Derivative) control algorithm, executed by a microcontroller, which provides dynamic stability and precise maneuverability—even in harsh or wind-deflected environments. This enables smooth, controlled flight over long distances and under varying meteorological conditions.
In another aspect of the invention, the drone is configured with six essential sensors, including the MQ2 gas leakage sensor, Raindrop sensor, MQ135 air quality sensor, and the DHT11 temperature and humidity sensor, among others, all orchestrated by a central microprocessor. An LCD panel is integrated into the system to display real-time sensor data throughout flight operations. The drone emphasizes a simplified yet robust assembly process, enabling fast deployment and maintenance. During final assembly, code is executed via a microcontroller cable, initiating the data collection process and activating real-time data display on the LCD.
In another aspect of the invention, unlimited and secure data transmission, the invention integrates a custom communication network architecture built on MQTT protocols and Node-RED servers, allowing both offline (ad-hoc) and internet-enabled modes of operation. This architecture ensures reliable, scalable, and uninterrupted data flow, even in disconnected environments, through a personalized broker and subscriber system. The drone is capable of autonomously connecting to a base station, where environmental data is uploaded and passed through advanced processing layers.
For forecasting and deep data analysis, the system includes an on-board computer that encodes sensor readings as vectors, processes them locally using lightweight AI models, and stores them for further downstream analysis. Upon base station synchronization, data is forwarded through a hybrid CNN and LSTM architecture, enabling real-time prediction, pattern detection, and trend analysis. A data analytics engine follows, decoding the information and extracting deeper insights. A dashboard interface completes the system, providing a user-friendly portal for full control, live monitoring, and visualization of all environmental metrics.
Thus the system is combined with Node-RED with a custom server infrastructure, enabling stable and intelligent drone operation via PID control, supporting unlimited and secure data collection, and performing in-depth environmental forecasting and analytics—all within a system that operates in any environment, whether connected or autonomous.
Brief Description of the Drawing:
The following figures can be used to gain a thorough grasp of the system and methodology of the current invention:

Fig 100- Detailed workflow and segregation of our UAV and proposed system
Fig 200- Detailed overview of our connection onboard our UAV
Fig 300- Customized propeller design for extra stability.

Figure 100 illustrates the integrated architecture of the environmental meteorological drone, highlighting the complete data flow from flight stabilization and sensor-based data acquisition to processing, forecasting, and real-time visualization.
Figure 200 presents a detailed block diagram representing the hardware architecture of the environmental meteorological drone, showing the interconnected components responsible for sensor interfacing, data processing, flight control, and power management.
Figure 300 illustrates a detailed orthographic top-view of a custom-designed four-blade propeller engineered for enhanced aerodynamic stability under diverse environmental conditions.

Detailed description of the embodiment
The following provides a detailed description of the embodiments of the disclosure as illustrated in the accompanying drawings. These embodiments are presented with sufficient detail to clearly convey the disclosure while encompassing all potential modifications, equivalents, and alternatives that fall within the scope of the appended claims. The accompanying drawings, which form an integral part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain its underlying principles. The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate an embodiment of the invention, and together with the description, serve to explain the principles of the invention. As used in the description herein and throughout the claims that follow, the meaning of “a,” “an,” and “the” includes plural reference unless the context dictates otherwise.
In the preferred embodiment, Figure 100 illustrates the overall operational flow of the environmental meteorological drone system, which is designed to collect, process, and analyze environmental data in real time under diverse and challenging weather conditions.
The drone is constructed using a precision-tuned aerodynamic frame, integrated with the DG1 propulsion engine, custom-designed propellers for flight accuracy, and a secure assembly of key components such as the flight controller, GPS module, and an array of cutting-edge environmental sensors. In an aspect of the invention , the central to the flight control system is a specialized PID (Proportional-Integral-Derivative) control algorithm, executed by a microcontroller, which provides dynamic stability and precise maneuverability—even in harsh or wind-deflected environments. This enables smooth, controlled flight over long distances and under varying meteorological conditions.
In another aspect of the invention, the drone is configured with six essential sensors, including the MQ2 gas leakage sensor, Raindrop sensor, MQ135 air quality sensor, and the DHT11 temperature and humidity sensor, among others, all orchestrated by a central microprocessor. An LCD panel is integrated into the system to display real-time sensor data throughout flight operations. The drone emphasizes a simplified yet robust assembly process, enabling fast deployment and maintenance. During final assembly, code is executed via a microcontroller cable, initiating the data collection process and activating real-time data display on the LCD.
In another aspect of the invention, unlimited and secure data transmission, the invention integrates a custom communication network architecture built on MQTT protocols and Node-RED servers, allowing both offline (ad-hoc) and internet-enabled modes of operation. This architecture ensures reliable, scalable, and uninterrupted data flow, even in disconnected environments, through a personalized broker and subscriber system. The drone is capable of autonomously connecting to a base station, where environmental data is uploaded and passed through advanced processing layers.
For forecasting and deep data analysis, the system includes an on-board computer that encodes sensor readings as vectors, processes them locally using lightweight AI models, and stores them for further downstream analysis. Upon base station synchronization, data is forwarded through a hybrid CNN and LSTM architecture, enabling real-time prediction, pattern detection, and trend analysis. A data analytics engine follows, decoding the information and extracting deeper insights. A dashboard interface completes the system, providing a user-friendly portal.
The top of the diagram outlines the hardware segment, starting with a high-stability flight controller powered by a specialized PID control algorithm, implemented via an on-board microcontroller. This algorithm ensures precise flight behaviour, even in turbulent conditions, enabling the drone to maintain position and orientation essential for accurate data acquisition.
Following the flight stabilization block, the system integrates a comprehensive set of environmental sensors—including gas sensors MQ2, MQ135, temperature and humidity sensors DHT11 , ACEBOTT Raindrop Sensor,etc.These sensors are routed through a central microcontroller, which synchronizes sensor inputs and ensures accurate, noise-reduced data collection.
Once collected, environmental data is forwarded to the on-board computer, where it is converted into encoded vector representations. These vectors are prepared for transmission and further analysis. When the drone comes within range of the base station, it automatically establishes a secure connection, enabling data offload without user intervention.
The software segment begins at this point. The data pipeline routes collected vectors through deep learning-based models, specifically CNN and LSTM algorithms, for forecasting weather trends and environmental shifts. A subsequent data analytics model evaluates tensor data for hidden patterns, anomalies, and deep structural relationships within the environmental datasets.
Finally, the processed data undergoes ad-hoc processing for lightweight inference, supporting low-latency applications. Results are made accessible through a dual-layer system: a personal broker–subscriber module for offline or decentralized usage, and a web-based visualization dashboard for real-time monitoring, user control, and system diagnostics.
In the preferred embodiment, seamless integration of hardware and software, real-time stability and data acquisition, autonomous communication, secure and scalable data handling, and intelligent forecasting and analysis in both connected and disconnected environments.
Figure 200 illustrates the embedded hardware architecture of the environmental meteorological drone system. At the core of the design is a central microcontroller, which acts as the interface and control hub for all peripheral modules. Connected to the microcontroller are multiple environmental sensors, responsible for measuring parameters such as temperature, humidity, air quality, and gas concentrations. These sensors are shown in three branches at the bottom of the figure, emphasizing the modular and expandable nature of the sensing layer.
The used sensors are:
DHT11 – Temperature and Humidity Sensor
• Function: Measures ambient temperature and relative humidity.
• Temperature Range: 0°C to 50°C
• Temperature Accuracy: ±2°C
• Humidity Range: 20% RH to 90% RH
• Humidity Accuracy: ±5% RH
• Output Signal: Digital (single-wire serial interface)
• Resolution:
o Temperature: 1°C
o Humidity: 1% RH
• Response Time: <5 seconds
• Units:
o Temperature in °C
o Humidity in % Relative Humidity (%RH)
MQ135 – Air Quality Sensor
• Function: Detects a wide range of gases such as ammonia (NH₃), nitrogen oxides (NOx), alcohol, benzene, smoke, and carbon dioxide (CO₂).
• Detection Range:
o NH₃: 10–300 ppm
o CO₂: 10–1000 ppm (approximate sensitivity)
o Benzene: 10–1000 ppm
• Sensitivity: High sensitivity to NH₃, NOx, alcohol, benzene, smoke
• Output Signal: Analog (voltage varies with gas concentration)
• Preheat Time: 24–48 hours for best accuracy
• Units: Parts Per Million (ppm)
– Flammable Gas Sensor
• Function: Detects flammable gases like LPG, methane (CH₄), propane, hydrogen (H₂), smoke.
• Detection Range:
o LPG: 200–10000 ppm
o Methane: 300–10000 ppm
o Hydrogen: 200–10000 ppm
• Response Time: <10 seconds
• Output Signal: Analog (voltage change with gas concentration)
• Units: ppm (Parts Per Million)

ACEBOTT Raindrop Sensor
• Function: Measures presence and intensity of rainfall (precipitation).
• Detection Mechanism: Analog voltage varies based on the amount of water on the sensor plate; more water = lower resistance = lower voltage. Also includes digital output based on a threshold.
• Detection Range: Qualitative (Dry, Drizzle, Moderate Rain, Heavy Rain based on analog voltage)
• Output Signal:
o Analog output: Proportional to amount of rainfall
o Digital output: Threshold-triggered (via onboard potentiometer)
• Units: No direct unit (mapped to rainfall levels by user calibration), typically interpreted in terms like mm/hr when calibrated against actual rain data.
SEN0456 – Barometric Pressure Sensor
• Function: Measures atmospheric (barometric) pressure and temperature.
• Sensor Chip: BMP280 (commonly used in SEN0456)
• Pressure Range: 300 hPa to 1100 hPa (equivalent to altitudes from ~-500m to 9000m)
• Pressure Accuracy: ±1 hPa
• Temperature Range: -40°C to 85°C
• Temperature Accuracy: ±1°C
• Output Signal: I2C or SPI (Digital)
• Units:
o Pressure: hectopascals (hPa) or millibar (mbar)
o Temperature: °C

Power is supplied through a dedicated power module and battery unit, ensuring reliable operation under varying flight conditions. The power path flows into both the microcontroller and the flight controller, which are responsible for real-time drone stability and navigation. The flight controller is assisted by a PID (Proportional–Integral–Derivative) control algorithm, shown as a dedicated module, which ensures adaptive and stable flight behavior in dynamic and turbulent environments. GPS functionality is integrated and fed directly into the flight controller to support geolocation, altitude hold, and waypoint tracking.
To support data-intensive applications, an on-board computer is connected to the microcontroller. This subsystem performs vector encoding and lightweight processing of sensor data. A dashed connection is used to indicate the optional interface with a dedicated server, which becomes active once the drone auto-connects to a base station for forecasting and advanced analytics using CNN and LSTM models.
The architecture reflects a distributed yet coordinated hardware-software interaction, where the microcontroller ensures low-level sensor data acquisition, the on-board computer handles computation and encoding, and the flight controller manages aerial dynamics. Together, they enable reliable operation, real-time environmental data collection, and intelligent forecasting in both connected and remote environments.
Figure 300 illustrates a detailed two-view technical drawing of a custom-designed drone propeller, highlighting both top and side profiles. The top view illustrates the overall span of the propeller as 250.00 -260.00 mm, providing a balanced thrust-to-diameter ratio ideal for medium-to-large drone platforms. The design features an aerodynamic S-curve blade profile, optimized for laminar airflow and reduced acoustic signature during operation. At the center of the blade is a mounting hub with a diameter of 11.85-13.10 mm, precisely machined for compatibility with standard motor shaft adaptors. The central bore diameter of 3.80-5.10 mm facilitates secure attachment to the motor axis. This bore is aligned concentrically within the mounting hub to minimize wobble and ensure rotational stability at high RPMs. The side view provides insight into the propeller’s aerodynamic cross-section and structural thickness. The maximum vertical height, or blade pitch thickness, is measured at 2.80-3.80 mm. This curvature is configured to support lift generation while resisting deformation under varying air pressure and temperature conditions, which are commonly encountered during flight in dynamic environmental monitoring missions. The asymmetric air foil geometry, visible in the side elevation, contributes to enhanced thrust and improved efficiency in forward flight. The shape also aids in maintaining consistent angular velocity across altitudes, which is critical for accurate sensor data acquisition and flight stability. Overall, the propeller in Figure 300 embodies a blend of aerodynamic optimization, precision engineering, and structural compatibility, aligning with the invention’s objectives for stable, energy-efficient, and accurate drone performance in diverse environmental conditions.
Algorithms used in the system is explained as follows:
Adaptive PID with Kalman Filter in a Drone Flight Controller
• Kalman Filter: Uses existing flight controller sensors like accelerometer, gyroscope, and barometer to filter noise and estimate the drone’s state.
• Adaptive PID: Adjusts the PID gains (Kp, Ki , Kd) based on the variance of the sensor data. Higher variance indicates unstable conditions requiring adjustment.
• Control Loop: Computes control signals using the PID controller and applies them to the drone’s state to achieve the desired target.

• The code uses the Kalman filter to estimate the state of the drone using noisy sensor data. • PID gains are dynamically adjusted based on sensor noise variance to ensure stable control under varying conditions.
• The control loop is modular and can be adapted for real-world implementation with standard drone sensors.

Novel Multi-Stage CNN-BiLSTM-Attention Architecture for Environmental Forecasting
Multi-Scale Spatio-Temporal Feature Extraction:
The architecture integrates parallel 1D convolutional branches with varying kernel sizes (3, 5, 7) to capture local environmental fluctuations across multiple temporal scales, followed by a Bidirectional LSTM for learning long-range dependencies in both forward and backward directions
.Dynamic Contextual Forecasting via Self-Attention:
A custom self-attention layer processes the BiLSTM outputs to assign dynamic importance to each time step, allowing the model to adaptively weigh historical sensor readings and generate precise multi-parameter environmental forecasts for real-time drone-based decision-making.
Advantage of the invention:
The advantage of the present invention is to develop an advanced environmental drone system that seamlessly integrates hardware-level stability with intelligent data acquisition and processing capabilities.
Further advantage of the present invention is to enable the drone to operate stably in dynamic and rough environmental conditions using a microcontroller-based flight controller running a specialized PID control algorithm.
Another advantage is to incorporate a wide array of environmental sensors (such as temperature, humidity, and air quality sensors), managed by a microcontroller, to ensure real-time, high-resolution data collection across varying terrains and climates.
Further advantage is to integrate an on-board computer capable of encoding environmental data as vectors, performing lightweight inference using embedded models, and preparing data for secure transmission and downstream analysis.
Yet further advantage of the invention is to provide automatic synchronization with a base station, where the collected data undergoes advanced processing through a hybrid CNN-LSTM architecture for predictive modeling, followed by in-depth analytics for pattern recognition and trend analysis.
Further advantage of the present invention is to establish a dual-mode communication architecture, comprising both an offline ad-hoc network (with a personalized broker and subscriber model) and an internet-enabled mode, thus ensuring uninterrupted data transmission and system control.
Further advantage is to enable real-time visualization and full control of the drone’s data and operation through an interactive dashboard, facilitating field deployment, monitoring, and decision-making in critical environmental research and applications.
It will be understood that the invention may be carried out into practice by skilled persons with many modifications, variations and adaptations without departing from its spirit or exceeding the scope of the claims in describing the invention for illustration.
Any inclusion to or deletion from the embodiment occurred, the specification is herein deemed as modified thus fulfilling the written description of all elements used in the claims so appended
, Claims:1. An environment Meteorological Drone for Comprehensive Environmental Data Collection (100-300) and Analysis characterizing by
-precision-tuned aerodynamic frame integrated with the DG1 propulsion engine which is configured propellers for flight accuracy and a secure assembly of key components of flight controller with GPS module wherein the central to the flight control system is a specialized PID (Proportional-Integral-Derivative) control algorithm, executed by a microcontroller, which provides dynamic stability and precise maneuverability—wherein in harsh or wind-deflected environments and enables smooth, controlled flight over long distances and under varying meteorological conditions.
- an array of cutting-edge plurality of environmental sensors embedded with MQ2 gas leakage sensor, MQ135 air quality sensor, and the DHT11 temperature and humidity sensor, ACEBOTT Raindrop Sensor , Flammable Gas Sensor
-an LCD panel is integrated into the system to display real-time sensor data throughout flight operations. The drone emphasizes a simplified yet robust assembly process, enabling fast deployment and maintenance
- built on MQTT protocols and Node-RED servers, allowing both offline (ad-hoc) and internet-enabled modes of operation. This architecture ensures reliable, scalable, and uninterrupted data flow, even in disconnected environments, through a personalized broker and subscriber system
-an on-board computer is connected to the microcontroller and the subsystem performs vector encoding and lightweight processing of sensor data to support data-intensive applications, wherein a dashed connection is used to indicate the optional interface with a dedicated server, which becomes active once the drone auto-connects to a base station for forecasting and advanced analytics using CNN and LSTM models and position and orientation essential for accurate data acquisition.

- power module and battery unit for power is supplied through a dedicated, ensuring reliable operation under varying flight conditions wherein the power path flows into both the microcontroller and the flight controller, which are responsible for real-time drone stability and navigation wherein the flight controller is assisted by a PID (Proportional–Integral–Derivative) control algorithm, shown as a dedicated module, which ensures adaptive and stable flight behavior in dynamic and turbulent environments and GPS functionality is integrated and fed directly into the flight controller to support geolocation, altitude hold, and waypoint tracking.
Wherein the system is characterised with an on-board computer that encodes sensor readings as vectors, processes it locally using AI models, and stores them for further downstream analysis wherein base station synchronization data is forwarded through a hybrid CNN and LSTM architecture, enabling real-time prediction, pattern detection, and trend analysis wherein a data analytics engine is used to decode the information and extracting deeper insights and a dashboard interface completes the system, providing a user-friendly portal wherein the system includes an on-board computer that encodes sensor readings as vectors, processes them locally using AI models, and stores them for further downstream analysis and after base station synchronization, data is forwarded through a hybrid CNN and LSTM architecture, enabling real-time prediction, pattern detection, and trend analysis for forecasting and deep data analysis.
2.The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein the drone is configured in the range of overall span of the propeller 250.00 -260.00 mm, providing a balanced thrust-to-diameter ratio ideal for medium-to-large drone platforms wherein the design features an aerodynamic S-curve blade profile which is optimized for laminar airflow and reduced acoustic signature during operation wherein at the center of the blade is a mounting hub with a diameter of 11.85-13.10 mm, precisely configured for compatibility with standard motor shaft adaptors and the central bore diameter of 3.80-5.10 mm and facilitates secure attachment to the motor axis wherein the bore is aligned concentrically within the mounting hub to minimize wobble and ensure rotational stability at high RPMs and the maximum vertical height and blade pitch thickness, is measured at 2.80-3.80 mm. wherein this curvature is configured to support lift generation and resisting deformation under varying air pressure and temperature conditions, which are commonly encountered during flight in dynamic environmental monitoring missions and the curvature is configured to support lift generation while resisting deformation under varying air pressure and temperature conditions, which are commonly encountered during flight in dynamic environmental monitoring missions.
3. The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein the DHT11 is embedded in the system for temperature and humidity Sensing and measures ambient temperature and relative humidity and the sensor temperature Range varies 0°C to 50°C, wherein temperature Accuracy: ±2°C, humidity Range: 20% RH to 90% RH, Humidity Accuracy: ±5% RH, Output Signal: Digital (single-wire serial interface) wherein resolution for temperature: 1°C and Humidity: 1% RH Response Time: <5 seconds
4. The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein the MQ135 – Air Quality Sensor detects a wide range of gases such as ammonia (NH₃), nitrogen oxides (NOx), alcohol, benzene, smoke, and carbon dioxide (CO₂) wherein detection range varies -NH₃: 10–300 ppm, CO₂: 10–1000 ppm (approximate sensitivity), Benzene: 10–1000 ppm and Output Signal is Analog wherein voltage varies with gas concentration and Preheat Time: 24–48 hours for best accuracy.
5.The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein the Flammable Gas Sensor detects flammable gases like LPG, methane (CH₄), propane, hydrogen (H₂), smoke. Wherein detection Range: LPG: 200–10000 ppm, Methane: 300–10000 ppm, Hydrogen: 200–10000 ppm, Response Time: <10 seconds Output Signal: Analog wherein voltage changes with gas concentration.
6. The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein ACEBOTT Raindrop Sensor measures presence and intensity of rainfall (precipitation). Wherein detection Mechanism is Analog and the voltage varies based on the amount of water on the sensor plate; and also configured with digital output based on a threshold, Detection Range: Qualitative (Dry, Drizzle, Moderate Rain, Heavy Rain based on analog voltage), Output Signal: is Analog output: Proportional to amount of rainfall and Digital output: Threshold-triggered (via onboard potentiometer)
7.The environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 1 wherein SEN0456 –detects Barometric Pressure Sensor measures atmospheric (barometric) pressure and temperature wherein Sensor Chip: BMP280 Pressure Range: 300 hPa to 1100 hPa (equivalent to altitudes from ~-500m to 9000m) Pressure Accuracy: ±1 hPa Temperature Range: -40°C to 85°C, Temperature Accuracy: ±1°C ,Output Signal: I2C or SPI (Digital)
8. The process of environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis is characterised by
- Stabilisation of the drone using flight stabilization block
-Collection of environmental data using the integrated environmental sensors—including gas sensors MQ2, MQ135, temperature and humidity sensors DHT11 , ACEBOTT Raindrop Sensor,
-. Synchronizing the sensor inputs and ensures accurate, noise-reduced data collection are routed through a central microcontroller, which send the collected, environmental data to the on-board computer, where it is converted into encoded vector representations.
-Forecasting and weather trends in environmental condition the software segment begins at this point wherein the data pipeline routes collected vectors through deep learning-based models, , for forecasting weather trends and environmental shifts wherein a data analytics model evaluates tensor data for hidden patterns, anomalies, and deep structural relationships within the environmental datasets.
-Processing the data after ad-hoc processing for lightweight inference, supporting low-latency applications and results are made accessible through a dual-layer system: a personal broker–subscriber module for offline or decentralized usage, and a web-based visualization dashboard for real-time monitoring, user control, and system diagnostics.
-Collecting vectors through deep learning-based models using CNN and LSTM algorithms, for forecasting weather trends and environmental shifts.
9. The process of environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 8 wherein Adaptive PID with Kalman Filter in a Drone Flight Controller uses existing flight controller sensors like accelerometer, gyroscope, and barometer to filter noise and estimate the drone’s state and Adaptive PID adjusts the PID gains based on the variance of the sensor data and wherein higher variance indicates unstable conditions requiring adjustment and the control loop computes control signals using the PID controller and applies it to the drone’s state to achieve the desired target wherein the microcontroller ensures low-level sensor data acquisition, the on-board computer handles computation and encoding, and the flight controller manages aerial dynamics wherein the code uses the Kalman filter to estimate the state of the drone using noisy sensor data and PID gains are dynamically adjusted based on sensor noise variance to ensure stable control under varying conditions wherein the control loop is modular and can be adapted for real-world implementation with standard drone sensors.
10. The process of environment Meteorological Drone for Comprehensive Environmental Data Collection and Analysis as claimed in claim 8 wherein the flight control system is a specialized PID (Proportional-Integral-Derivative) control algorithm, executed by a microcontroller, which provides dynamic stability and precise maneuverability wherein in harsh or wind-deflected environments and enables smooth, controlled flight over long distances and under varying meteorological conditions and it provides a balanced thrust-to-diameter ratio ideal for medium-to-large drone platforms wherein the design features an aerodynamic S-curve blade profile which is optimized for laminar airflow and reduced acoustic signature during operation wherein at the center of the blade is a mounting hub wherein the curvature is configured to support lift generation and resisting deformation under varying air pressure and temperature conditions, which are commonly encountered during flight in dynamic environmental monitoring missions and the curvature is configured to support lift generation while resisting deformation under varying air pressure and temperature conditions, which are commonly encountered during flight in dynamic environmental monitoring missions.