Description:SYSTEM TO DETERMINE WEIGHT OF AN ELECTRIC VEHICLE
Field of the Invention
[0001] The present invention relates to determination of weight pertinent to an electric vehicle (EV). More specifically, the present invention relates to a system based on a architectural paradigm of one or more functional elements to precisely estimate/ detect weight of EV, based on a plurality of relevant parameters.
Background
[0002] The background description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.
[0003] In consideration of the rapid advancement of current logistics industry, the problem of transport vehicles overloading has become graver. Overloaded vehicles cause extensive damage to the road and huge economic losses to society, which seriously threaten traffic safety. The overloading of transport vehicles has become one of the crucial issues of traffic management. In the field of overload detection, traditional static weighing system needs to intercept a part of vehicles from the normal traffic flow and weigh them in a static field; it takes a long time and affects the normal traffic. There are also many unsolved problems in the dynamic weighing technology, such as requirement of large infrastructure, time consuming, manual weighing etc., which is inconvenient for installation and maintenance. Vehicle weight limits are exceeded due to inadvertent or deliberate weight added to the design feature for a vehicle. Presently, there are no easily installed sensor devices or methodologies that enable such determination. In commercial applications where vehicle up fitters start with a bare chassis, add vehicle structure and then allow the ultimate operator/user to add additional weight such as tools or materials, the vehicle can easily be overloaded in excess of the manufacturers intended weight limits. Such overloading produces both unsafe operating conditions as well as contributing to vehicular wear and tear in excess of its intended use.
[0004] One or more problems that are exemplary associated vehicle overload are enumerated herein. For instance, if a vehicle is overloaded then parts of the vehicle (such as a wagon, electric car, truck or trailer and the like) are strained and various mechanical systems can break down. Thus, vehicle reliability is inversely proportional to any overloaded conditions that may occur. Further, if the vehicles are overloaded they do not handle particularly well. If for instance, one considers a truck, and if the right side of the truck has 1,000 pounds more than the left side, then if the boom operator is extending in that direction, he will be exceeding the safety limits of the truck. Thus, weight distribution and thus truck levelling can involve a safety issue.
[0005] Moreover, when vehicle is manufactured, they are given a gross vehicle weight rating that does not take into account any additional equipment that is added to a vehicle. By way of example, customers will start to put tools and equipment on a newly bought or leased truck and may for instance fill up utility bays with equipment. However, once the truck leaves the manufacturing factory, there is no way to easily ascertain whether the loaded truck is within the maximum gross vehicle weight rating assigned to the truck, or for instance how the truck is maintained with respect to the rating. While the trucks may leave the factory in a compliant condition they are essentially shipped as an empty truck. Typically utility operators decide what is going to be carried on a truck with no particular thought to the final weight of the loaded truck. In short, there is no convenient way to give a truck operator an immediate understanding that his vehicle is overloaded or that uneven loading or a load shift has occurred which may result in tipping or unsafe operation.
[0006] Therefore, in light of the detailed analysis of the aforementioned facts, a system henceforth is proposed to control and manage one or more functional elements architecturally arranged to determine a weight of an electric vehicle (EV). The scope of the proposed system cannot be restricted to the present invention and can stretch beyond the limitations of the present invention, through further course of research and development. Nonetheless, the proposed system is open to modification and necessary updates, as and when required, in accordance with the embodiments of present disclosure.

Objects of the Invention
[0007] An object of the present disclosure is to overcome one or more drawbacks associated with conventional mechanisms.
[0008] An object of the proposed system can be the automation for the control and management of architectural setup associated with the weighing of electric vehicle (EV).
[0009] An objective of the present disclosure is to enable remote monitoring of EV weight, without weight sensor and/or off-road weighing station or pressure sensitive strips laid on or in the roadway.
[00010] An object of the proposed system can be a transmission of an alert about a vehicle overload to a concerned user/authoritarian.
[00011] An object of the proposed system can be the penalization of violators who breach or intend to breach the policies and rules associated with freight and logistics.
[00012] An object of the proposed system can be to render systemic integration with one or more types of electric vehicle (such as car, bike, 3 wheeler, truck, bus, train and the like).
[00013] An object of the proposed system can be the communicative linking of the user to the plurality of appliance, to manage and control them remotely.
[00014] An object of the proposed system can be to render an alternative to the conventional control and management scheme employed for determining the weight of an electric vehicle.
[00015] An object of the proposed system can be the inclusion of systemic and systematic practices which can be employed to effectively control and manage one or more operational functions associated with the determination of EV weight.
[00016] An object of the proposed system can be compatibility and portability.
[00017] An object of the proposed system can be a self-learning so a system aging related issues would have little or no effect on the weight estimations of EV.
[00018] An object of the proposed system can be to determine a real-time vehicle weight which can be used even for a high frequency algorithm such as emergency brakes, emergency reverse and other safety controls.
[00019] An object of the proposed system can be innovation of a conventional architecture deployed for detection of vehicle overload.
[00020] An object of the proposed system is to amplify computational speed by utilizing Machine Learning engine.
Summary
[00021] The present invention relates to determination of weight pertinent to an electric vehicle (EV). More specifically, the present invention relates to a system based on a state-of-the-art architectural paradigm of one or more functional elements to precisely estimate/ detect weight of EV, based on a plurality of relevant parameters.
[00022] The following presents a simplified summary of various aspects of this disclosure in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements nor delineate the scope of such aspects. Its purpose is to present some concepts of this disclosure in a simplified form as a prelude to the more detailed description that is presented later.
[00023] The following paragraphs provide additional support for the claims of the subject application.
[00024] In an aspect, the present invention provides a system to determine a weight of an electric vehicle (EV), the system comprises: an electric propulsion unit comprising: an electric motor is arranged to generate a driving force to propel the EV; a sensing kit comprising: a motor speed sensor is arranged to determine a current revolution per minute (RPM) number of the electric motor; a power measurement sensor is arranged to measure power parameter of the motor; and a speed sensor is arranged to determine a current speed of the EV; a vehicle control unit (VCU) comprising: a data storage memory is arranged to store a vehicle weight database (VWD) that comprising pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and a microcontroller unit (MCU) is arranged to: receive the determined RPM number, the measured power parameter and the determined speed; filter the VWD based on the received RPM number, the power parameter, the speed to create a filter data; apply a machine learning technique to the created filtered data to generate multiple weight prediction models; select a best prediction model from the generated multiple weight prediction models; and apply the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
[00025] In another aspect, the present invention provides a method for determining a weight of an electric vehicle (EV), the method comprises: utilizing an electric propulsion unit comprising an electric motor and a sensing kit, wherein the electric motor is arranged to generate a driving force to propel the EV; arranging the sensing kit for: determining, through a motor speed sensor, a current revolution per minute (RPM) number of the electric motor; measuring, through a power measurement sensor, power parameter of the motor; and determining, through a speed sensor, a current speed of the EV; utilizing a vehicle control unit (VCU) comprising data storage memory for storing a vehicle weight database (VWD) that comprises pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and utilizing a microcontroller unit (MCU) is arranged for: receiving, the determined RPM number, the measured power parameter and the determined speed; filtering, the VWD based on the received RPM number, the power parameter, the speed to create a filter data; applying, a machine learning technique to the created filtered data to generate multiple weight prediction models; selecting, a best prediction model from the generated multiple weight prediction models; and applying, the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
[00026] In an embodiment, the sensing kit comprises a gyro sensor to determine a current inclination angle of the EV.
[00027] In an embodiment, the MCU filter the created filter dataset based on the determined current inclination angle to create a second filter dataset.
[00028] In an embodiment, the MCU create a list of secondary weight prediction model based on the created second filtered dataset.
[00029] In an embodiment, the MCU predict a secondary weight based on the created secondary weight prediction model.
[00030] In an embodiment, the MCU calculates an optimum air pressure condition of tyre, if the predict weight is greater than a threshold limit.
[00031] In an embodiment, the MCU calculates a safety setting based on the predict weight.
[00032] In an embodiment, the MCU calculates an aging factor based on the predicted weight.

Brief Description of the Drawings
[00033] The features and advantages of the present disclosure would be more clearly understood from the following description taken in conjunction with the accompanying drawings in which:
[00034] FIG. 1 illustrates an architectural paradigm of a system to determine a weight of an electric vehicle (EV). The term “electric vehicle” or “EV” as used herein and may be throughout a course of present disclosure.
[00035] FIG. 2 illustrates an architectural paradigm of the VCU, in accordance with the embodiments of present disclosure.
[00036] FIG. 3 represents an architectural arrangement of one or more functional unit, which can be executed collectively or individually or selectively or sequentially by the microcontroller unit of the VCU, associated with the system, in accordance with the embodiments of present disclosure.
[00037] FIG. 4 represents a portray a complete workflow or block diagram of the proposed system.
[00038] Fig. 5 illustrate exemplarily steps for determining a weight of an electric vehicle (EV), in accordance with embodiments of the present disclosure.

Detailed Description
In the following detailed description of the invention, reference is made to the accompanying drawings that form a part hereof, and in which is shown, by way of illustration, specific embodiments in which the invention may be practiced. In the drawings, like numerals describe substantially similar components throughout the several views. These embodiments are described in sufficient detail to claim those skilled in the art to practice the invention. Other embodiments may be utilized and structural, logical, and electrical changes may be made without departing from the scope of the present invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims and equivalents thereof.
The use of the terms “a” and “an” and “the” and “at least one” and similar referents in the context of describing the invention (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The use of the term “at least one” followed by a list of one or more items (for example, “at least one of A and B”) is to be construed to mean one item selected from the listed items (A or B) or any combination of two or more of the listed items (A and B), unless otherwise indicated herein or clearly contradicted by context. The terms “comprising,” “having,” “including,” and “containing” are to be construed as open-ended terms (i.e., meaning “including, but not limited to,”) unless otherwise noted. Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the invention.
The present invention relates to determination of weight pertinent to an electric vehicle (EV). More specifically, the present invention relates to a system based on a state-of-the-art architectural paradigm of one or more functional elements to precisely estimate/ detect weight of EV, based on a plurality of relevant parameters.
According to an illustration made in figure 1, showcasing an architectural paradigm of a system 100 to determine a weight of an electric vehicle (EV) 112. The term “electric vehicle” or “EV” as used herein and may be throughout a course of present disclosure, can relate to, but not restricted to a vehicle such as a car, bicycle, scooter, wheelchair, truck, bus, train, boat, ship, and other known variants, architecturally enabled to functionally operate on an electrical energy source. However, throughout the course of detailed description and one or more relevant drawing, an electrical car or a fuel cell car or a series hybrid car or a parallel hybrid car may be emphasized as an epitome of EV 112.
In an embodiment, the system 100 may comprise an electric propulsion unit 102, a vehicle control unit (VCU) 104 and other known vehicle components. The system 100 may be architecturally integrated with the EV 112 wherein EV 112 may further incorporate a power source 106, an electrical circuitry 108, and a charging interface 110. A person ordinarily skilled in art would prefer that the aforesaid one or more functional elements of the system 100, may be functionally or operatively interlinked with each other (such as via a network interface, may not be restricted to Bluetooth, Wi-Fi and the like or through an electrical or an electronic arrangement).
Referring to the preceding embodiment, one or more functional elements of the system 100 may be arranged as an assembly of functional elements (collectively such as an apparatus or device) or separately (individually such as an embodiment/engine/aspect or module) or a combination thereof, whereby de-limiting the scope of architectural arrangement of the system 100. The pictorial representation made in figure 1, can be considered as a mere depiction or a demonstration of system 100, thus cannot limit a scope of the present invention. However, to those ordinarily skilled or extra ordinarily skilled in art may prefer that one or more functional elements/embodiments included in the architectural setup of system 100, can be modified and updated, as and when necessary, in accordance with the embodiments of present disclosure.
In an embodiment, the electric propulsion unit 102 may be arranged to generate a driving force to propel the EV 112 and sense one or more relevant parameters, pertinent to the weight determination of EV 112. The electric propulsion unit 102 may comprise an electric motor 102a and a sensing kit 102b and. The term “electrical propulsion” as used herein may relate to, but not limited to an employment of electrical power or electrical energy to drive or electrically propel the EV 112. For instance, one or more elements of the electric propulsion unit 102 can be purely electrical, that may not employ any combustion engine or a generated energy therein due to a combustion of fossil fuel, to drive the EV 112 (such as an electric car). A data transceiver may be arranged to transmit of one or more information from the electric propulsion unit 102, over an exemplary network interface.
In an embodiment, the electric motor 102a may be arranged to generate a driving force (such as a torque or a rotating force generated therein) to propel the EV 112. The electric motor 102a may be selected from a Permanent Magnet Synchronous Motor (PMSM), an axial flux motor, a three Phase AC Induction Motor, a wound-rotor motor and other known examples thereof. The term “electric motor” as implicated herein, may relate to, but not restricted to a device or means which can be adapted or enabled to convert an electrical energy or input into a mechanical work or output and can cause the propulsion of EV 112, in accordance with the embodiments of present disclosure.
In an embodiment, the sensing kit 102b may be arranged to detect or sense one or more relevant parameters pertinent to EV 112, can be associated with one or more objective or purpose or a motive of deploying system 100. The sensing kit 102b may comprise a motor speed sensor 102b1, a power measurement sensor 102b2, a speed sensor 102b3 and other known elements thereof. For instance, a gyro sensor may be arranged to determine a current inclination angle (such as a measurement of, not limited to a tilt or a lateral orientation) of the EV 112.
In an embodiment, the motor speed sensor 102b1 can be arranged to determine a current revolution per minute (RPM) number for the electric motor 102a. For instance, the motor speed sensor 102b1 may be selected from not limited to a digital tachometer, vibration sensor, a proximity sensor, a rotary torque sensor and other examples thereof. A person ordinarily skilled in art may prefer that the motor speed sensor 102b may be adapted or enabled (such as by transforming or converting a commutation frequency (such as an electrical power or electrical energy cycles or pulses may be correlated to a speed of electric motor 102a) to RPM or converting an angular motion or relative position of a shaft into one or more pulse signal) to precisely or accurately detect/determine or sense a current revolution per minute (RPM) or angular speed number for the electric motor 102a.
In an embodiment, the power measurement sensor 102b2 can be arranged to measure power parameter of the electric motor 102a. For instance, the power measurement sensor 102b2 may employ an exemplary formula such as, not limited to P = t * ?, (wherein P may relate to a power parameter, t may indicate a torque or a driving force and ? may indicate an angular speed or RPM) to measure power parameter of the electric motor 102a. Alternatively, the power measurement sensor 102b2 can measure amount of electric energy subjected to the electric motor 102a from a battery pack. The power measurement sensor 102b2 can be selected from voltmeter, ammeter and any other variant.
In an embodiment, the speed sensor 102b3 can be arranged to determine a current speed of the EV 112. For instance, the speed sensor 102b3 can be a device that can be enabled to measure current speed (such as a measure of fastness or slowness of movement or a rate of covering a distance per unit time, can be represented as Km/hour or miles per hour and the like) of the EV 112. The speed sensor 102b3 can measure the current speed of the EV 112, may be based on a transmission/transaxle output or a wheel speed. The speed sensor 102b3 may comprise a silicon integrated circuit (IC) in a sensor head, which can be hermetically sealed with an over molded plastic. Alternatively, speed of EV 112 can be determined based on change in location (which is tracked using GPS sensor) in a unit time.
The VCU 104 may receive, in runtime or polling manner, the measured measure power parameter (from power measurement sensor 102b2), current speed (from speed sensor 102b3) and current RPM number (from motor speed sensor 102b1), through wireless (e.g., Bluetooth, WiFi etc.) or wired network (e.g., controller area network (CAN) bus, Local Interconnect Network (LIN) bus, FlexRay bus, Media Oriented Systems Transport (MOST) bus and combination thereof). It would be appreciated that the present disclosure is not limited to any type of data communication protocol and a person ordinarily skill in the art can use appropriate data communication protocol. The VCU 104 may deploy artificial intelligence-based state of art prediction functionality to determine current weight of EV 112. The determined current weight can be communicated to a driver, remote server, manufacturer, insurance provider, a fleet manager, governmental agency and like that. Thus, the present disclosure enables real-time weight determination irrespective of location and without requirement of weighbridge facility. Further, present disclosure can be used for all kinds of vehicles, irrespective of vehicle type (e.g., truck, bus, scooter, car) and minimum weight of EV 112. Further, the real time determination of weight can also be used to prevent fraudulent activity (reporting of lower weight) by driver, fleet manager and other parties for various purposes like warranty claiming, road tax or insurance premium calculation. Moreover, VCU 104 may fine tune the determine weight based on the determined current inclination angle (determined through gyro sensor). As vehicle moves on upward hill, power supply to electric motor 102a would be high, RPM number would be less (as higher torque would require for driving uphill) and speed would be lower (in comparison to flat road driving). Weight prediction during drive on uphill would be higher (compared to actual or current weight). Similarly, if EV 112 is driving on downhill, predicted weight (without consideration of inclination angle) would be lower (as speed, rpm would be higher). Thus, if inclination angle (e.g., road slope condition) is not considered while evaluating the weight of EV, predicted weight would have lower precision and accuracy, and higher error. Thus, fine tuning of predicted weight based on the information of gyro sensor improves overall functionality. In case EV 112 is driving on relative flat condition, then there would not be fine tuning.
According to an illustration made in figure 2, may pictorially depict an architectural paradigm of the VCU 104, in accordance with the embodiments of present disclosure. In an embodiment, the vehicle control unit (VCU) 104 may be architecturally configured/ enabled with one or more elements, such as but not limited to a data storage memory 104a, a microcontroller unit 104b (MCU) and other known examples thereof. For instance, the VCU 104 may be arranged to perform one or more functional operations such as a storage or cache of one or more relevant information critical to the deployment of system 100. Further, the VCU 104 can be adapted to execute one or more operational/functional instruction pertinent to an acquisition one or more data critical to one or more objective of system 100, a data filtration, a data processing/ analysis, a digital output and other known examples thereof.
Referring to an epitome of preceding embodiment, the VCU 104 may be enabled with a bus which may allow a communication among one or more functional embodiments of the control unit or sensor(s) or other electronic components. For instance, one or more processor or microprocessor or microcontroller, a memory or storage device deployed thereof. The VCU 104 may be enabled with a processing module or engine which may include one or more processor (such as a central processing unit, a graphics processing unit, an accelerated processing unit and the like), one or more microprocessor, and/or another type of processing component (such as a field-programmable gate array (“FPGA”), an application-specific integrated circuit (“ASIC”), and the like) that can interpret and/or execute instructions.
As utilized herein, or throughout a course of present disclosure, the term such as a “control,” a “system,” an “interface,” and the like can be intended to refer to a computer-related entity, either a hardware, software (e.g., in execution), and/or firmware and other known examples thereof. For example, a control unit can be a process running on a processor, a processor, an object, an executable, a program, and/or a computer. By a way of illustration, both an application running on a server and the server can be a control unit.
In an embodiment, the data storage memory 104a can be arranged to store a vehicle weight database (VWD) wherein VWD may comprise one or more pre-stored weight values. Each of the weight value can be tagged with a plurality of velocity values, a plurality of RPM values, a multiple power data and multiple inclination angle information. The term “data storage memory” as used herein can be construed as, not limited to a computer storage medium which can be included or integrated in a computer-readable storage device, a computer-readable storage substrate, a random or serial access memory array or device, or a combination of one or more of them. The data storage memory 104a may include a random-access memory (“RAM”), a read only memory (“ROM”), and/or another type of dynamic or static storage device (e.g., a flash, magnetic, or optical memory) that may store information and/or instructions for an execution by the processor. A person ordinarily skilled in art that one or more values (such as weight, velocity and the like) may be arranged in a plurality of rows and columns which can enable to discern an understanding of one or more values from each other (such as a dependence or independence of a value on other values, respectively). In an alternative embodiment, the VWD can be stored in a dedicated computing device such as high-speed computer etc. The VCU 104 may retrieve or access the VWD, which is arranged at remote location, through wireless or wired network. In another embodiment, VCU 104 may comprise an independent processor (e.g., machine learning (ML) engine 104c) to access VWD that can be stored locally or at remote location. The separate processor may increase computing speed and reduce computational load on VCU 104. Alternatively, ML engine 104c and/or MCU 104b can fine-tune the received or generated models to improvise accuracy of weight prediction.
Referring to the preceding embodiment, the one or more pre-stored weight values may be sorted (may be according to date, a time, vehicle type, an ascending or descending order of weight and other known examples thereof) in VWD. The VWD may be ramified into plurality of sections, wherein each section may arrange one or more relevant information, associated with one or more parameter information and the like. The term “tagged” as used herein may relate to, but not restricted to labelled, marked, an identifier, stickered, stamped and other known examples thereof. To those ordinarily skilled in art may prefer that one or more information may be arranged chronologically in a VWD which can be stored for any duration, and accessed or modified as and when necessary, in accordance with the embodiments of present disclosure.
In an embodiment, the microcontroller unit 104b (MCU) or ML engine 104c may be arranged to perform one or more functional operation associated with the determination of the weight of the electric vehicle 112. For instance, one or more functional operation may be inclusive a reception of parameter information, a filtration of VWD, an application or implementation of machine learning technique, a selection and application of a best prediction model to determine the weight of the EV 112. In an alternate embodiment, the MCU 104b can also be enabled to calculate an electrical need of the electric motor 102a and can provide an additional high energy boost to meet a surge in energy or power demand (may be due to an acceleration and other known causes thereof) to propel the EV 112.
According to an illustration made in figure 3, may portray an architectural arrangement 200 of one or more functional unit, which can be executed collectively or individually or selectively or sequentially by the microcontroller unit 104b or ML engine 104c of the VCU 104, associated with the system 100, in accordance with the embodiments of present disclosure. Dual processing system of VCU 104 would be advantageous over solitary processor, which perform various functions (e.g., weight prediction, control of BMS, communication etc.) of EV 112. The architecture 200 may comprise a data acquisition unit 202, a data filtration unit 204, a ML model-based calculation unit 206, and other known elements thereof. One or more executable routines may be stored in the data storage memory 104a wherein each of the executable routine, may include one or more functional unit as mentioned herein.
In an embodiment, the data acquisition unit 202 may be arranged to receive the determined RPM number, the measured power parameter and the determined speed, may be via a data transceiver arranged with the sensing kit 102b, over an exemplary network interface (such as Bluetooth, WI-FI and the like). To those ordinarily skilled in art may prefer that one or more received information may be stored for any duration in the data storage memory 104a configured with the VCU 104 and may be accessed, modified and updated as and when necessary, in accordance with the embodiments of present disclosure. Alternatively, the acquisition unit 202 may receive data (e.g., firmware update package, previously developed models, VWD data etc.) from external computer.
In an embodiment, the data filtration unit 204 may be executed by MCU 104b or ML engine 104c to filter the VWD based on the received RPM number, the power parameter and the current speed to create a filter data. To filter WED, data filtration unit 204 may use pre-set criteria (e.g., ± 10 % deviation for RPM, ± 8 to 20 % variation for speed, ± 5 to 30% of range of received power parameter) to create filter data. A person ordinarily skilled in art may prefer that one or more information may be completely segregated in the filtered data, from the VWD, post an execution or implementation of the data filtration unit 204. Further, the MCU 104b can filter the created filter dataset based on the determined current inclination angle to create a second filter dataset.
Referring to the preceding embodiment, the MCU 104b can create a list of secondary weight prediction model based on the created second filtered dataset. A person ordinarily skilled in art would prefer to extract one or more information chronologically or serially sorted, associated with one or more filtering criteria such as received RPM number, the power parameter and the current speed through an application of a data filter (such as through an implementation of a toggle filter which may toggle down one or more filtering criteria or a radio filter to select a filter and the like). An implementation of the data filtration unit 204 may generate a consistent or more stable and noise free dataset in the created filtered data. The implementation of data filtration unit 204 may improve an efficiency of system 100 deployed to determine the weight of the EV 112.
In an embodiment, the ML model-based calculation unit 206 may be executed by the MCU 104b to apply a machine learning technique to the created filtered data to generate multiple weight prediction models. To those skilled in art may prefer a plurality of machine learning technique which can be selectively or preferably applied for the generation of multiple weight prediction models. One or more machine learning technique may be implicated, which can selected from an artificial neural network (ANN), a Linear Regression, Gaussian Process, Ensemble Boosting Tree with Ada Boost, Ensemble Boosted Tree, support vector machine (SVM) and Ensemble Bagging Tree with Random Forest and other known examples thereof. For instance, an execution of the ML model-based calculation unit 206 may ramify information with the created filtered data into a training set (~70-75%), a test set (~10-15%) and a validation set (~70-75%). The training set may be deployed to develop one or more machine learning model to compare one or more received information associated with such as received RPM number, the power parameter, the current speed, the corresponding weight values and the like, pre stored in the VWD. In another embodiment, high-speed computing device can be deployed to generate weight prediction models. The high-speed computing device can use data of VWD to adjust weight assignment to individual parameters (e.g., speed, RPM etc.) to finetune the models. The fine tune model(s) can be provided to MCU 104b or ML engine 104c for prediction of vehicle weight. Alternatively, the VCU 104 may transmit the measured data to cloud server, which can use machine learning based weight prediction. The cloud server may access VWD to generate weight prediction models using suitable machine learning techniques (e.g., SVM, recursive neural network etc.). It would be appreciated that processing unit of cloud server and/or high-speed computing device can perform all steps of ML model-based calculation unit 206.
Referring to the preceding embodiment, a machine learning algorithm may learn and train from a training data and can find a relationship or data pattern, develop an understanding, make one or more decisions, and can develop a confidence. The performance of each of developed multiple weight prediction model can be evaluated by using a test set. Similarly, the validation set can be deployed by the MCU 104b to cross-validate and verify one or more evaluation or assessment performed on the test set. The validation set may be used to evaluate one or more filtered information (such as received RPM number, the power parameter and the current speed, corresponding pre stored weight values and the like) of the developed machine learning model. Similarly, model generation unit 206 may apply similar machine learning technique (which used previously for filter data set) or different machine learning technique on second filter dataset to create secondary weight prediction models. This secondary weight prediction model can be used to incorporate inclination angle information to the predict a secondary weight. Thus, the secondary weight prediction models based on correction in predicted weight improve overall weight prediction functionalities of present disclosure.
Still referring to the preceding embodiment, a model application unit may be arranged to select a best prediction model from the generated multiple weight prediction models. For instance, the mathematical model having higher classification ability (e.g., lower false prediction) can be selected as best prediction models from the generated multiple weight prediction models. Further, the selected prediction model may be applied to the received RPM number, the power parameter, and the current speed to predict weight of the EV 112. The best prediction model may be selected based on root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE), Supervised Learning, Unsupervised Learning, Reinforcement Learning, Prediction Speed and Training Time.
The MCU 104b may predict a secondary weight based on the created secondary weight prediction model.
In an alternate embodiment, an input such as a motor power, motor speed and a gyroscope data (which may be stored in data storage memory 104a of the VCU 104) may be utilized find a relation between a given input and an actual data. Further the MCU 104b may generate a prediction function that can predict the vehicle weight by taking one or more real-time sample into account. The prediction function may be coded in a machine learning (ML) Engine which can be installed onboard VCU 104. For instance, an ANN algorithm may be employed to generate a predictive function to create a relation between the input and the actual vehicle weight. An error detection and correction factor can adjust one or more parameters in accordance with the embodiments of present disclosure, wherein one or more parameters may be referred to as ANN weights and biases. The relationship relation between the given input and an actual data may become more and more efficient by correcting the relation in multiple iterations. A final trained predictive function may have error nearly equal to zero, hence can provide a precise vehicle weight estimations.
Referring to the preceding embodiment and an illustration made in figure 4, may portray a complete workflow or block diagram of the proposed system 100. The block diagram may showcase one or more operational steps which may be sequentially performed in order to meet a plurality of motive and objectives of the present invention. One or more relevant or critical data associated with the weight determination of EV 112, may undergo through the one or more operational steps or processes (such as data acquisition, data organization/sorting, ML simulation, development of ML training model, a coding of ML training model into ML engine, a weight determination), can efficiently emphasize an ambit of the present invention. As illustrated the present disclosure can enable considering of weight being towed such as a weight of a trailer with material present within the tailer. To include towed weight, conventional vehicle weight determination system requires specific sensor unit (which needs to be mount on towed weight) or weighbridge needs to be redesign to accommodate complete towed tailor. However, such implementation is expensive and require deployment of dedicated infrastructure. As present disclosure can utilize speed, inclination angle, RPM number and power parameters to predict weight, in comparison to conventional weight sensor-based method which omit consideration of towed weight for weight prediction.
As illustrated in Fig. 4, in step 1 to 3, the EV 102 undergoes test phase to collect data by varying speed, inclination angle, motor input/output, and other relevant parameters. It would be appreciated that the test phase can be performed on actual vehicle drive on road or can be mounted on dynamometer. Alternatively, the test data can be collected based on computation simulation tool such as MATLAB/Simulink. The test data can be collected for specific model of EV 112. The model specific test data can be stored in VWD, the model specific VWD would be advantageous to improve overall efficiency of the weight prediction process of instant disclosure. The generated test data would be used for generation of ML weight prediction models using machine learning technique. The model generation can be performed at high-speed computing device. As described, the data of VWD can be split into training set (which would be used to generate models), and test set (which would be used for validation of generated model). A robust prediction model can be identified based on prediction accuracy and, lower false positive/negative rate. For efficient model generation and/or training of ML models, following can be used. The various data of VWD can be normalized or transformed, using following formula:
x^'=(( x - x_min ))/(( x_max - x_min ) ) (1)

y^' = log(1+y) (2)

Where x’ is the normalized input, x is the actual input, x¬min is minimum input value, xmax is the maximum input value, y’ is transformed output and y is actual output. It would be appreciated that data normalization and/or transformation can be carried out for various parameters like motor speed, motor power, and inclination angle.
Value input for the activation function can be computed based on following equation by using,
Y_in=b+?_(i=1)^n¦?x_i w_i ? (3)

Where Yin is the Input value of an activation function, n will be the total number of input data used, wi is the neural network weight parameter value, xi is the input value, and b is the bias.
Weight assignment to individual parameters can be updated by,
w_(i(new))= w_(i(old))+ x_i (y-Y_in)? (4)

Where w¬i(new) is the new weight value, w¬i(old) is the previous weight value, xi is the input value at instant i, y is the actual output, Yin ¬is the input value of an activation function and ? is the learning rate, wherein bias values will be updated by,
b_((new))= b_((old) )+(y-Y_in)? (5)

Where b(new) is the new bias value and b(old) is the previous bias value.
The generated ML models can undergo for evaluation phase using test set to recognize best model, which can be used for weight prediction of EV 112. The best model can be coded in VCU 104 or MCU 104b or ML-Engine 104b. The MCU 104b or ML engine 104c can utilize the data collected through sensing kit 102b, to predict weight of EV 112. The predicted weight can be displayed on a display screen that is associated with EV 112 or can be communicated to relevant party (e.g., manufacturer, insurance provider etc.)
In an embodiment, the predicted weight (which can include weight being hauled in the vehicle and/or weight being towed as a trailer) can be monitored by manufacturer or warranty provider or insurance company to determine rate/frequency of overload condition (e.g., predicted weight greater than manufacturer's specifications).
As overall all weights on the EV 112 (along with self weight) can have impact on mechanical and structural integrity and performance of EV 112 and/or various subcomponents thereof. Frequent or continuous overload weight condition (e.g., predicted weight is greater than a threshold value such as 10-15%, >20%, >35% and like that from a manufacture specified weight) may create a plethora of mechanical and structural problems that may significantly impact the operation, aging, performance and other imperative performance parameters of EV 112.
The MCU 104b or cloud server (of manufacturer, insurance company etc.) can access historic predicted weight information to calculate an aging factor. The aging factor can be used for various purposes such as determination of health of condition of EV 112, insurance premium, resale value, pollution tax etc.
In an alternative embodiment, MCU 104b can be arranged to calculate an optimum air pressure condition (e.g., air pressure of each tyre ), upon detection of overload weight condition (e.g., the predicted weight is greater than a threshold limit from a manufacturer specified weight). Further, MCU 104b can be arranged to transmit an alert (e.g., SMS, email, push notification) to a computing device (e.g., smartphone, laptop etc.), which can be accessed by driver, manager, service engineer etc. The prompt notification can result in reduction of any relevant parameter (e.g., speed, RPM, power parameter) can reduce overall effect of weight on EV 112. Thus, present disclosure can improve lifespan of EV 112 and also reduce maintenance cost (which would require against repairing of wear and tear due to overload condition).
Further, MCU 104b can be configured to calculate a safety

Claims
I/We Claim:
1. A system to determine a weight of an electric vehicle (EV), the system comprises:
an electric propulsion unit comprising:
an electric motor is arranged to generate a driving force to propel the EV;
a sensing kit comprising:
a motor speed sensor is arranged to determine a current revolution per minute (RPM) number of the electric motor;
a power measurement sensor is arranged to measure power parameter of the motor; and
a speed sensor is arranged to determine a current speed of the EV;
a vehicle control unit (VCU) comprising:
a data storage memory is arranged to store a vehicle weight database (VWD) that comprising pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and
a microcontroller unit (MCU) is arranged to:
receive the determined RPM number, the measured power parameter and the determined speed;
filter the VWD based on the received RPM number, the power parameter, the speed to create a filter data;
apply a machine learning technique to the created filtered data to generate multiple weight prediction models;
select a best prediction model from the generated multiple weight prediction models; and
apply the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
2. The system as claimed in claim 1, wherein the sensing kit comprises a gyro sensor to determine a current inclination angle of the EV.
3. The system as claimed in claim 2, wherein the MCU filter the created filter dataset based on the determined current inclination angle to create a second filter dataset.
4. The system as claimed in claim 3, wherein the MCU create a list of secondary weight prediction model based on the created second filtered dataset.
5. The system as claimed in claim 4, wherein the MCU predict a secondary weight based on the created secondary weight prediction model.
6. The system as claimed in claim 1, wherein the MCU calculates an optimum tyre air pressure condition, if the predict weight is greater than a threshold limit.
7. The system as claimed in claim 1, wherein the MCU calculates a safety setting based on the predict weight.
8. The system as claimed in claim 1, wherein the MCU calculates an aging factor based on the predicted weight.
9. A method for determining a weight of an electric vehicle (EV), the method comprises:
utilizing an electric propulsion unit comprising an electric motor and a sensing kit, wherein the electric motor is arranged to generate a driving force to propel the EV;
arranging the sensing kit for:
determining, through a motor speed sensor, a current revolution per minute (RPM) number of the electric motor;
measuring, through a power measurement sensor, power parameter of the motor; and
determining, through a speed sensor, a current speed of the EV;
utilizing a vehicle control unit (VCU) comprising data storage memory for storing a vehicle weight database (VWD) that comprises pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and
utilizing a microcontroller unit (MCU) is arranged for:
receiving, the determined RPM number, the measured power parameter and the determined speed;
filtering, the VWD based on the received RPM number, the power parameter, the speed to create a filter data;
applying, a machine learning technique to the created filtered data to generate multiple weight prediction models;
selecting, a best prediction model from the generated multiple weight prediction models; and
applying, the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
10. The method as claimed in claim 9, wherein the MCU calculates an aging factor based on the predicted weight.

SYSTEM TO DETERMINE WEIGHT OF AN ELECTRIC VEHICLE
Abstract
The proposed system deploys a plurality of functional embodiments to precisely determine a weight of an electric vehicle (EV). The proposed system deploys an electric propulsion unit whereby an electric motor and a sensing kit are arranged to generate a driving force to propel the EV and detect one or more relevant parameters, respectively, pertinent to one or more objective and purpose thereof. A vehicle control unit comprises a data storage memory and a microcontroller unit. The data storage memory holds a vehicle weight database while the microcontroller unit is arranged to perform one or more functional operations associated with the proposed system, emphasizing an ambit of present invention. , Claims:Claims
I/We Claim:
1. A system to determine a weight of an electric vehicle (EV), the system comprises:
an electric propulsion unit comprising:
an electric motor is arranged to generate a driving force to propel the EV;
a sensing kit comprising:
a motor speed sensor is arranged to determine a current revolution per minute (RPM) number of the electric motor;
a power measurement sensor is arranged to measure power parameter of the motor; and
a speed sensor is arranged to determine a current speed of the EV;
a vehicle control unit (VCU) comprising:
a data storage memory is arranged to store a vehicle weight database (VWD) that comprising pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and
a microcontroller unit (MCU) is arranged to:
receive the determined RPM number, the measured power parameter and the determined speed;
filter the VWD based on the received RPM number, the power parameter, the speed to create a filter data;
apply a machine learning technique to the created filtered data to generate multiple weight prediction models;
select a best prediction model from the generated multiple weight prediction models; and
apply the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
2. The system as claimed in claim 1, wherein the sensing kit comprises a gyro sensor to determine a current inclination angle of the EV.
3. The system as claimed in claim 2, wherein the MCU filter the created filter dataset based on the determined current inclination angle to create a second filter dataset.
4. The system as claimed in claim 3, wherein the MCU create a list of secondary weight prediction model based on the created second filtered dataset.
5. The system as claimed in claim 4, wherein the MCU predict a secondary weight based on the created secondary weight prediction model.
6. The system as claimed in claim 1, wherein the MCU calculates an optimum tyre air pressure condition, if the predict weight is greater than a threshold limit.
7. The system as claimed in claim 1, wherein the MCU calculates a safety setting based on the predict weight.
8. The system as claimed in claim 1, wherein the MCU calculates an aging factor based on the predicted weight.
9. A method for determining a weight of an electric vehicle (EV), the method comprises:
utilizing an electric propulsion unit comprising an electric motor and a sensing kit, wherein the electric motor is arranged to generate a driving force to propel the EV;
arranging the sensing kit for:
determining, through a motor speed sensor, a current revolution per minute (RPM) number of the electric motor;
measuring, through a power measurement sensor, power parameter of the motor; and
determining, through a speed sensor, a current speed of the EV;
utilizing a vehicle control unit (VCU) comprising data storage memory for storing a vehicle weight database (VWD) that comprises pre-stored weight values, wherein the each of the weight value is tagged with multiple velocity values, multiple RPM values, multiple a power data and multiple inclination angle information; and
utilizing a microcontroller unit (MCU) is arranged for:
receiving, the determined RPM number, the measured power parameter and the determined speed;
filtering, the VWD based on the received RPM number, the power parameter, the speed to create a filter data;
applying, a machine learning technique to the created filtered data to generate multiple weight prediction models;
selecting, a best prediction model from the generated multiple weight prediction models; and
applying, the selected best prediction model to the received RPM number, the power parameter, the speed to predict weight of the EV.
10. The method as claimed in claim 9, wherein the MCU calculates an aging factor based on the predicted weight.