Estimation of inclination angle of a vehicle

Abstract

Examples for estimation of inclination angle of a vehicle are described. In an example, an estimation model is trained for estimation of inclination angle of vehicle without the use of an IMU sensor. During implementation, once the estimation model is trained based on training information, an input vehicle characteristic is obtained which may indicate motional changes occurring in a vehicle while traveling on the road. These input vehicle characteristics may be processed based on the estimation model to determine a weightage parameter corresponding to the input vehicle characteristic.

Description

ESTIMATION OF INCLINATION ANGLE OF A VEHICLEBACKGROUND

[0001] Generally, inclination measurement in various applications, such as automotive and mobile devices rely on Inertial Measurement Unit (IMU) sensors to determine orientation and movement. The IMU sensors typically provide accelerometer data, which is then used to calculate inclination angles. An IMU sensor generally combines accelerometers, gyroscopes, and sometimes magnetometers to provide motion tracking capabilities. Accelerometers within an IMU measure linear acceleration along one or more axes, while gyroscopes detect rotational motion, and magnetometers assess orientation relative to the Earth's magnetic field. Thus, IMU sensors are sophisticated devices that combine multiple sensors to detect changes in linear acceleration, angular rate, and magnetic field surrounding the vehicle. By processing these signals, IMUs can provide real-time data on the vehicle's orientation, including its inclination angle.BRIEF DESCRIPTION OF FIGURES

[0002] Systems and / or methods, in accordance with examples of the present subject matter are now described and with reference to the accompanying figures, in which:

[0003] FIG. 1 illustrates an environment comprising an estimation system for estimating inclination angle of a vehicle, as per an example of the present subject matter;

[0004] FIG. 2 illustrates a computing system for estimating inclination angle of a vehicle, as per another example of the present subject matter;

[0005] FIG. 3 illustrates a computing system for training an estimation model for estimating inclination angle of a vehicle, as per an example of the present subject matter; and

[0006] FIG. 4 illustrates a method for training for estimating inclination angle of a vehicle, as per an example of the present subject matter.DETAILED DESCRIPTION

[0007] Generally, automotives utilize IMU sensor(s) to determine orientation and movement. The data from the IMU sensors can be processed to determine the orientation of an object in three-dimensional space, commonly referred to as an inclination angle. The inclination angle of a vehicle, often referred to as a pitch angle, is a measure of an angle between a vehicle's longitudinal axis and a horizontal plane. The inclination angle is a pivotal parameter in understanding a vehicle's orientation and is particularly relevant when navigating uneven terrains, such as hills or ramps. The accuracy and reliability of data captured by the IMU sensors is therefore of utmost concern, as any deviation can lead to errors in the system's output and performance.

[0008] Despite the widespread use of the IMU sensors, the IMU sensors face challenges due to high sensitivity to external factors, such as vibrations, shocks, and thermal variations. These factors may introduce noise into the accelerometer data, leading to fluctuations that may compromise the accuracy of the calculated inclination angles. The precision of inclination measurements is thus contingent on the stability and quality of the IMU sensor data, which can be affected by environmental and operational conditions.

[0009] Approaches for reducing dependency on potentially noisy IMU sensor data and instead utilizing alternative, more reliable parameters, such as motor torque and throttle value for calculating the inclination angle, are described. The approaches of the present subject matter aim to mitigate the impact of disturbances on inclination measurements, thereby enhancing an overall reliability and performance of systems that require precise orientation data.

[0010] The present subject matter describes systems and methods for estimation of the inclination angle of an automotive without the use of anIMU sensor are described. It may be noted that the present examples described in the context of automotives may include, but not limited to, vehicles such as two-wheelers, three-wheelers, four-wheelers such as electric cars, scooters, trucks, vans, utility vehicles, etc.

[0011] In an example, the system may comprise an estimation engine for training an estimation model for estimating an inclination angle of an automotive without the use of an IMU sensor.

[0012] Once the estimation model is trained, an input vehicle characteristic is obtained from the vehicle which may indicate motional changes occurring in a vehicle while traveling on the road. These input vehicle characteristics may be processed by the estimation model to determine a weightage parameter corresponding to the input vehicle characteristic.

[0013] Further, the estimation model may calculate the inclination angle based on the determined weightage parameter using equation (1 ) provided below,Inclination Angle = (Wi*(d(Motor_Torque) / dt) + W2*(d(Throttle) / dt) + W3*(d(Vehicle speed) / dt) + W4*Motor Torque + W5*Throttle + W6*Vehicle Speed )+C ....(1 )

[0014] Herein, ‘C’ is a constant for compensating error and offset, Wi and W4 are the weightage parameters determined corresponding to an input torque value, W2 and W5 are the weightage parameters determined corresponding to an input throttle value, and W3 and We are the weightage parameters corresponding to an input vehicle speed.

[0015] It is pertinent to note that the weightage parameters indicate the amount of contribution of each of the vehicle characteristics and their corresponding derivatives in calculating the inclination angle. Further, the estimation model is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

[0016] During training, the estimation model may be trained by obtaining a training vehicle characteristic and an actual inclination angle corresponding to the training vehicle characteristic. In an example, training data used for training the estimation model may comprise historical data relating to training vehicle characteristic corresponding to the training vehicle characteristic which may have been collected in the past based on the similar operating conditions of the vehicle and may also be used as training information for training estimation model.

[0017] The estimation model is trained based on the training information and weight data. In one example, the estimation model when trained may predict a weightage parameter corresponding to an input vehicle characteristic. The weight data corresponds to the weightage parameter associated with the input vehicle characteristic. Further, an inclination angle of the road may be calculated based on the determined weightage parameter.

[0018] The present disclosure offers several technical advantages over traditional inclination measurement systems that rely on Inertial Measurement Unit (IMU) sensors, such as enhanced accuracy, reduced sensitivity to disturbances, improved reliability, cost-effectiveness, operational insight, application versatility, enhanced vehicle safety and stability, and more. The examples of the present subject matter relies on a stable data namely, Motor Torque and Throttle, thus avoiding any faulty results due to fragile sensor outputs, and further helps with flexibility of placing the IMU sensor on the vehicle at positions prone to more degree of freedom.

[0019] Further, by utilizing a machine learning algorithm to determine the weightage of selected parameters, the present disclosure can adapt to various driving conditions and vehicle dynamics, leading to a more accurate calculation of the inclination angle. Further, unlike IMU sensors, which can be affected by noise and external vibrations, the disclosed examples rely on parameters that are inherently less sensitive to such disturbances, resultingin more stable inclination measurements. The data-driven approach ensures that the inclination angle is calculated based on real-time operational data, which enhances the reliability of the measurement under different load and road conditions.

[0020] As may be understood, eliminating the dependence on IMU sensors can reduce the overall cost of the system by avoiding the expense associated with these sensors and their maintenance. The use of machine learning allows for continuous improvement of the system as more data is collected, enabling the algorithm to refine its calculations for even greater precision over time. The present examples also provide insights into the vehicle’s operational state, such as load conditions and road type, which can be valuable for other vehicle control systems and driver assistance features, and can be applied to a wide range of vehicles and is not limited to those equipped with IMU sensors, thereby, broadening its applicability across the automotive industry. The disclosed examples can also improve the performance of these systems, potentially leading to increased safety for passengers.

[0021] The manner in which the example computing systems are implemented is explained in detail with respect to FIGS. 1 -4. While aspects of the described computing system may be implemented in any number of different electronic devices, environments, and / or implementations, the examples are described in the context of the following example device(s). It may be noted that drawings of the present subject matter shown here are for illustrative purposes and are not to be construed as limiting the scope of the claimed subject matter.

[0022] FIG. 1 illustrates an environment 100 comprising an estimation system for estimating inclination angle of a vehicle, for example, vehicle 120. The vehicle 120 may include vehicle hardware, such as an estimation system 102 (also referred to as system 102) which may be responsible for managing and / or controlling the input parameters for estimating inclination angle of the vehicle 120. In an example, the vehicle hardware of the vehicle120 may include hardware that is used to control the vehicle through real- world environments based on the sensor(s) data, one or more machine learning, neural network-based learning model, or deep learning models, or the like.

[0023] In an example, the system 102 may comprise processor(s) 104, memory(s) 106, instruction(s) 108, an estimation engine 1 10, an estimation model 1 12, training data 1 14, training information 1 16, and other data 1 18. The instructions 108 are fetched from memory 106 and executed by a processor 104 included within the system 102. The system 102 may be deployed within the cloud server of the vehicle 120. The estimation engine 1 10 may be implemented as a combination of hardware and programming, for example, programmable instructions to implement a variety of functionalities. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the estimation engine 1 10 may be executable instructions, such as instructions 108. Such instructions may be stored on a non-transitory machine-readable storage medium which may be coupled either directly with system 102 or indirectly (for example, through networked means). In an example, the estimation engine 1 10 may include a processing resource, for example, either a single processor or a combination of multiple processors, to execute such instructions. In the present examples, the non-transitory machine-readable storage medium may store instructions, such as instructions 108, that when executed by the processing resource, implement estimation engine 1 10. In other examples, the estimation engine 1 10 may be implemented as electronic circuitry.

[0024] Continuing further, the estimation model 1 12 is trained for estimation of inclination angle of the vehicle 120 without the use of an IMU sensor. Once the estimation model 1 12 is trained based on training information 116, an input vehicle characteristic is obtained which may indicate motional changes occurring in the vehicle 120 while traveling on the road, on an inclined plane having an inclination angle ‘6’ (shown in FIG.1 ). These input vehicle characteristics may be processed based on the estimation model 112 to determine a weightage parameter corresponding to the input vehicle characteristic. Although, the vehicle 120 is depicted as moving uphill, the vehicle 120 may also be moving downhill. In either of such situations, the current approaches may be utilized without deviating from the scope of the present subject matter. The inclined surface (as described in FIG. 1 ) may have varying gradient.

[0025] The vehicle characteristics may include, but are not limited to, a motor torque, a throttle position, a vehicle speed, or combination thereof. It may be noted that the vehicle characteristics may be obtained from a plurality of sensors installed onto the vehicle 120. In the present context, throttle Value depicts the torque requirement by a user, for example, the user riding the vehicle 120, and the more the throttle the more the torque requested by the user. Thus, this value is used to understand the value of torque requested. Motor Torque is used to determine the load applied on the vehicle, i.e., to clarify the reduction of speed for the same value of throttle is due to an increased load on the vehicle because of the pillion rider or weight or due to inclination angle. Motor torque helps determine the various other factors such as the Force applied by the vehicle. The vehicle speed is used to determine the acceleration or retardation of the vehicle 120, which helps in determining the vehicle's momentum.

[0026] By utilizing the above vehicle characteristics, throttle requirements of the user and the respective torque value may be determined, which may then be correlated with the vehicle speed for determining the load applied on the vehicle when the state of the vehicle is under high load or low load, for example, with respect to a pillion or without pillion or under uphill / downhill condition. Further, state of the vehicle 120, for example, if the vehicle 120 is running over a straight surface, ascending or descending can be determined using throttle value and vehicle speed. Thus, using the above characteristics, the entire state of the vehicle i.e., whether the vehicle is loaded / unloaded or over an inclination / straight roadcan be attained. The vehicle characteristics mentioned above may be obtained from a plurality of sensors installed onto the vehicle.

[0027] Further, the estimation model 112 is trained to calculate the inclination angle based on the determined weightage parameter using the following relation,Inclination Angle = (Wi*(d(Motor_Torque) / dt) + W2*(d(Throttle) / dt) + W3*(d(Vehicle speed) / dt) + W4*Motor Torque + W5*Throttle + W6*Vehicle Speed )+C - (1 )

[0028] Herein, ‘C’ is a constant for compensating error and offset, Wi and W4 are the weightage parameters determined corresponding to an input torque value, W2 and W5 are the weightage parameters determined corresponding to a input throttle value, and W3 and We are the weightage parameters corresponding to a input vehicle speed.

[0029] It is pertinent to note that the weightage parameters indicate the amount of contribution of each of the vehicle characteristics and their corresponding derivatives in calculating the inclination angle. Further, the estimation model 1 12 is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

[0030] The system 102 may further include training data 1 14 which pertains to the historical data which may have been collected in the past based on the similar operating conditions in the vehicle 120 and may be used as training data for training the estimation model 112. The estimation system 102 may be configured for installation within vehicle 120 and for use by the vehicle 120 for estimation of inclination angle of the vehicle 120 without the use of an IMU sensor.

[0031] FIG. 2 illustrates a computing system 200 for estimating inclination angle of a vehicle, for example vehicle 120 as shown in FIG. 1. Examples of system 200 include, but are not limited to, a portable computer, laptops, mobile phones, notebooks and other types of computing system. Although not depicted, the system 200 may include other components, suchas interfaces to communicate over the network or with external storage or computing devices, display, input / output interfaces, operating systems, applications, data, and the like, which have not been described for brevity. The system 200 may include a processor(s) 202 and a memory(s) 204. The processor 202 may be implemented as microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and / or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) 202 is configured to fetch and execute computer-readable instructions stored in memory(s) 204 for estimating inclination angle of the vehicle 120.

[0032] The system 200 may further include an estimation engine 206, same as estimation engine 1 10 of FIG. 1 which may be coupled to the processor(s) 202, and memory(s) 204. The estimation engine 206, amongst other functions, may obtain 208 an input vehicle characteristic indicating motional changes occurring in the vehicle 120 while travelling on a road. Further, the estimation engine 206 processes 210 the input vehicle characteristic based on an estimation model to determine a weightage parameter corresponding to the input vehicle characteristic, wherein the estimation model is trained based on a training vehicle characteristic and an actual inclination angle of the road corresponding to the training vehicle characteristic. Thereafter, the estimation engine 206 calculates an inclination angle of the road based on the determined weightage parameter.

[0033] FIG. 3 illustrates a computing system for training an estimation model for estimating inclination angle of a vehicle. The estimation model, for example, estimation model 312 (same as 1 12) is trained based on training information 308 which may be obtained prior to training of the estimation model 312.

[0034] In an example, the training system 302 may be communicatively coupled to a data repository 318 through a network 316. In another example, repository 318 may reside inside the system 102 as well. Therepository 318 may further include training data 320. In an example, the training data 320 comprises historical data relating to training vehicle characteristic 310 corresponding to the training vehicle characteristic which may have been collected in the past based on the similar operating conditions of the vehicle 120 and may also be used as training information 308 comprising the training vehicle characteristic 310 for training estimation model 312. The training system 302 may include a cloud-based deep learning infrastructure that may use artificial intelligence to analyze the input parameters for estimation of inclination angle of the vehicle 120 without the use of an IMU sensor.

[0035] The network 316 may be a private network or a public network and may be implemented as a wired network, a wireless network, or a combination of a wired and wireless network. The network 316 may also include a collection of individual networks, interconnected with each other and functioning as a single large network, such as the Internet. Examples of such individual networks include, but are not limited to, Global System for Mobile Communication (GSM) network, Universal Mobile Telecommunications System (UMTS) network, Personal Communications Service (PCS) network, Time Division Multiple Access (TDMA) network, Code Division Multiple Access (CDMA) network, Next Generation Network (NGN), Public Switched Telephone Network (PSTN), Long Term Evolution (LTE), and Integrated Services Digital Network (ISDN).

[0036] The training system 302 may further include a training engine 306. In an example, the instructions 304 are fetched from a memory and executed by a processor included within the training system 302. The training engine 306 may be implemented as a combination of hardware and programming, for example, programmable instructions to implement a variety of functionalities. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the training engine 306 may be executable instructions, such as instructions 304. Such instructions may bestored on a non-transitory machine-readable storage medium which may be coupled either directly with the training system 302 or indirectly (for example, through networked means). In an example, the training engine 306 may include a processing resource, for example, either a single processor or a combination of multiple processors, to execute such instructions. In the present examples, the non-transitory machine-readable storage medium may store instructions, such as instructions 304, that when executed by the processing resource, implement training engine 306. In other examples, the training engine 306 may be implemented as electronic circuitry.

[0037] The instructions 304, when executed by the processing resource, cause the training engine 306 to train an estimation model, such as an estimation model 312. The instructions 304 may be executed by the processing resource for training the estimation model 312 based on a set of training information 308 comprising training vehicle characteristic(s) 310. As also described in conjunction with FIG. 1 , training information 308 may be correlated by the training engine 306 of the training system 302 for determining the load applied on the vehicle when the state of the vehicle is under high load or low load, for example, with respect to a pillion or without pillion or under uphill / downhill condition. The correlation of training information 308 pertains to calculation of the inclination angle based on the determined weightage parameter as per the examples of the present subject matter. The estimation model 312 is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

[0038] FIGS. 4 illustrate example method 400 for training an estimation model, in accordance with examples of the present subject matter. The order in which the above-mentioned methods are described is not intended to be construed as a limitation, and some of the described method blocks may be combined in a different order to implement the methods, or alternative methods. During operation, the estimation engine 306, amongstother functions, may extract a plurality of input vehicle characteristic(s) associated with a vehicle, for example, a vehicle 120. Further, the above- mentioned methods may be implemented in suitable hardware, computer- readable instructions, or combination thereof. The steps of such methods may be performed by either a system under the instruction of machine executable instructions stored on a non-transitory computer readable medium or by dedicated hardware circuits, microcontrollers, or logic circuits. For example, the methods may be performed by a training system, such as training system 302.

[0039] In an implementation, method 400 may be performed under an “as a service” delivery model, where the training system 302, operated by a provider, receives programmable code. Herein, some examples are also intended to cover non-transitory computer readable medium, for example, digital data storage media, which are computer readable and encode computer-executable instructions, where said instructions perform some or all the steps of the above-mentioned methods. In an example, the method 400 may be implemented by the training system 302 for training the estimation model 312 based on training information 308 comprising training vehicle characteristic(s) 310.

[0040] At block 402, training data may be obtained. In one example, the training engine, for example training engine 306 may obtain training data 320 from a repository 318. In an example, the training data 320 comprises historical data relating to training vehicle characteristic corresponding to the training vehicle characteristic which may have been collected in the past based on the similar operating conditions of the vehicle 120 and may also be used as training information 308 for training estimation model 312.

[0041] At block 404, an estimation model is trained. For example, the training engine 306 may train the estimation model 312 based on training information 308, wherein the estimation model 312 when trained may predict a weightage parameter corresponding to an input vehicle characteristic, and wherein the weight data corresponds to the weightageparameter associated with the input vehicle characteristic. The weightage parameters indicate the amount of contribution of each of the vehicle characteristics and their corresponding derivatives in calculating the inclination angle. The weight data corresponds to the weightage parameter associated with the input vehicle characteristic.

[0042] At block 406, the values of the weightage parameter may be adjusted. For example, the training engine 306 may adjust the values of weightage parameters corresponding to each of the vehicle characteristics based on the difference between the estimated inclination angle and the actual inclination angle. The estimation model 312 is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

[0043] Although examples for the present disclosure have been described in language specific to structural features and / or methods, it is to be understood that the appended claims are not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed and explained as examples of the present disclosure.

Claims

I / We Claim:1 . A system comprising: a processor; and an estimation engine coupled to the processor, wherein the estimation engine is to: obtain an input vehicle characteristic indicating motional changes occurring in a vehicle while travelling on a road; process the input vehicle characteristic based on an estimation model to determine a weightage parameter corresponding to the input vehicle characteristic, wherein the estimation model is trained based on a training vehicle characteristic and an actual inclination angle of the road corresponding to the training vehicle characteristic; and calculate an inclination angle of the road based on the determined weightage parameter.

2. The system as claimed in claim 1 , wherein the vehicle characteristics is one of a motor torque, a throttle position, a vehicle speed, or combination thereof.

3. The system as claimed in claim 1 , wherein the vehicle characteristics are obtained from a plurality of sensors installed onto the vehicle.

4. The system as claimed in claim 1 , wherein the estimation engine is to calculate the inclination angle based on the determined weightage parameter using the following relation:Inclination Angle = (Wi*(d(Motor_Torque) / dt) + W2*(d(Throttle) / dt) + W3*(d(Vehicle speed) / dt) + W4*Motor Torque + W5*Throttle + We*Vehicle Speed )+C; wherein ‘C’ is a constant for compensating error and offset, Wi and W4 are the weightage parameters determined correspondingto an input torque value, W2 and W5 are the weightage parameters determined corresponding to a input throttle value, and W3 and We are the weightage parameters corresponding to a input vehicle speed.

5. The system as claimed in claim 1 , wherein the weightage parameters indicate amount of contribution of each of the vehicle characteristics and their corresponding derivatives in calculating the inclination angle.

6. The system as claimed in claim 1 , wherein the estimation model is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

7. A method comprising: obtaining training information comprising a training vehicle characteristic and an actual inclination angle corresponding to the training vehicle characteristic; and training an estimation model based on the training information and a weight data, wherein the estimation model when trained is to predict a weightage parameter corresponding to an input vehicle characteristic, and wherein the weight data corresponds to the weightage parameter associated with the input vehicle characteristic.

8. The method as claimed in claim 7, wherein the training vehicle characteristic is one of a motor torque, a throttle position, a vehicle speed, or combination thereof.

9. The method as claimed in claim 7, wherein the training information is obtained from a plurality of vehicles, wherein motor torque is obtained from a torque sensor, throttle position is obtained from a throttle position sensor, and the vehicle speed is obtained from a wheel speed sensor.

10. The method as claimed in claim 7, wherein an inclination angle corresponding to an input vehicle characteristic is determined using the following relation based on the determined weightage parameter:Inclination Angle =(Wi*(d(Motor_Torque) / dt) + W2*(d(Throttle) / dt) + W3*(d(Vehicle speed) / dt) + W4*Motor Torque + W5*Throttle + We*Vehicle Speed )+C; wherein ‘C’ is a constant for compensating error and offset, Wi and W4 are the weightage parameters determined corresponding to an input torque value, W2 and W5 are the weightage parameters determined corresponding to a input throttle value, and W3 and We are the weightage parameters corresponding to a input vehicle speed.1 1. The system as claimed in claim 10, wherein the method further comprises: adjusting the values of weightage parameters corresponding to each of the vehicle characteristics based on the difference between the estimated inclination angle and the actual inclination angle.

12. The method as claimed in claim 7, wherein the weightage parameters indicate amount of contribution of each of the vehicle characteristics and their corresponding derivatives in calculating the inclination angle.

13. The method as claimed in claim 7, wherein the estimation model is trained to output a confidence score along with the estimated inclination angle, indicating the reliability of the estimation.

Images