Description:Complete Specification:
The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed.

Field of the invention:
[0001] The present invention relates to a steering angle sensor unit for a steering wheel assembly of a vehicle and method of operation of the same.

Background of the invention:
[0002] Automotive is one such industry where sensors are essential components. The sensors transform physical quantities such as pressure or acceleration into outputs signals that could be used as an input for the controllers for data processing. Considering the volume of the vehicles sold and the competition in the industry, any cost advantage technique could be a life savior for the OEM and the suppliers.

[0003] According to a prior art WO12052688 method and system for redundancy of a measurement signal of a steering wheel angle sensor is disclosed. The subject of the present invention is a method and a system for redundancy of a measurement signal delivered by a sensor of the angle of a steering wheel of a vehicle. The aim of the invention is to allow the detection of defects of the angle sensor by effecting a redundancy between the main signal measured on the steering wheel and the signal reconstructed on the basis of information emanating from the vehicle, in a relatively short detection time. This redundancy allows the system to detect an error in the main measurement signal and to remedy this error. The invention proposes for this purpose a mathematical model linking the main signal with the other information emanating from the vehicle, such as the differential of the speed of the front or rear wheels, the rate of yaw and/or the lateral acceleration of the vehicle.

Brief description of the accompanying drawings:
[0004] An embodiment of the disclosure is described with reference to the following accompanying drawings,
[0005] Fig. 1 illustrates a block diagram of a steering angle sensor unit for a steering wheel assembly of a vehicle, according to an embodiment of the present invention, and
[0006] Fig. 2 illustrates a method of operation of the steering angle sensor unit for the steering wheel assembly of the vehicle, according to the present invention.

Detailed description of the embodiments:
[0007] Fig. 1 illustrates a block diagram of a steering angle sensor unit for a steering wheel assembly of a vehicle, according to an embodiment of the present invention. The steering angle sensor unit 102 comprises, an angle sensor 108 mounted to a steering column 106 of the steering wheel assembly 104, a controller 110 connected to the angle sensor 108 to determine a first steering parameter, characterized in that, the controller 110 configured to estimate a second steering parameter, through an estimation model 114 stored in a memory element 112 of the controller 110. The second steering parameter is redundant of the first steering parameter. The at least one of the first steering parameter and the second steering parameter is used for steering parameter-dependent vehicle functions. The steering parameter is at least one of the steering angle and a rack position. The steering wheel assembly 104 is either steer-by-wire (SbW) system or conventional system where rack and pinion based mechanism is used. In the conventional solutions, two values of steering parameters are provided using two physical angle sensors 108, however as per the present invention, the steering angle sensor unit 102 is a device which provides two values of the steering parameter, one is determined in real time using the angle sensor 108 and the other is estimated through the estimation model 114. The angle sensor 108 is shown at the end of the steering column 106, however the angle sensor 108 is possible to be positioned as per requirement and not limited to the same. Few examples of the vehicle functions are autonomous driving, active return of the steering, end stop of the steering, parking assist, and the like.

[0008] According to an embodiment of the present invention, the estimation model 114 comprises a physical model and a correction model. The physical model estimates the second steering parameter based on input parameters comprising yaw rate, lateral acceleration and wheel velocity, and the correction model comprises a Machine Learning/ Artificial Intelligence (ML/AI) based trained modules to correct the second steering parameter estimated by the physical model. The modules are trained using at least one of a Neural Networks, Random forest Models and the like. The training of the correction model is done during production/manufacturing stage before the vehicle 100 are rolled out for consumers. The training of the correction model is performed using the real-time value of the steering parameter captured either from the angle sensor 108 or auxiliary sensor (another sensor to measure the steering parameter) mounted solely for the training purposes.

[0009] According to an embodiment of the present invention, the second steering parameter is correctable with at least one of the trained modules and real-time values of the angle sensor 108 when the vehicle 100 is being driven. For example, if the vehicle 100 is being driven by a driver, and the steering angle sensor unit 102 is functioning well (without any errors), then the controller 110 is configured to use the real time determined value of the first steering parameter as measured by the angle sensor 108 for vehicle functions. The controller 110 also uses the same to train the and adapt the correction model. In other words, the correction model learns and adapts from the difference between the actual value sensed by the angle sensor 108 and the estimated value of the second steering parameter. Hence, if the angle sensor 108 faulters, then the controller 110 takes the second steering parameter as the main input for various vehicle functions till the angle sensor 108 is repaired or replaced or corrected.

[0010] According to an embodiment of the present invention, the controller 110 comprises any one of a first control unit and a combination of the first control unit and a second control unit. In the case of the first control unit, both the first steering parameter and the second steering parameter are determined and estimated by the controller 110 or the first control unit. In the case of the combination of the first control unit and the second control unit, the first control unit is interfaced with the angle sensor 108 to determine the first steering parameter and the second control unit is configured to estimate the second steering parameter.

[0011] According to an embodiment of the present invention, the controller 110 to estimate steering parameter for the steering wheel assembly 104 of the vehicle 100 is disclosed. The controller 110 configured to receive signals for input parameters from respective means, characterized in that, the controller 110 also configured to estimate the steering parameter, through the estimation model 114 stored in the memory element 112 of the controller 110. The estimation model 114 comprises the physical model and the correction model, and the input parameters are selected from yaw rate, lateral acceleration, and wheel velocity. The respective means corresponds to sensors to measure the yaw rate, lateral acceleration and wheel velocity or derived using models in dependence of other physical parameters of the vehicle 100.

[0012] The physical model estimates the second steering angle based on input parameters comprising yaw rate, lateral acceleration and wheel velocity, and the correction model comprises the Machine Learning/ Artificial Intelligence (ML/AI) based trained modules to correct the second steering parameter estimated by the physical model. The training of the correction model is done during production/manufacturing stage before the vehicle 100 are rolled out for consumers. The training of the correction model is performed using the real-time value of the steering parameter determined either from the angle sensor 108 or auxiliary sensor (another sensor to measure the angle) mounted solely for the training/learning purposes.

[0013] According to the present invention, the controller 110 is provided with necessary signal detection, acquisition, and processing circuits. The controller 110 is a control unit which comprises memory element 112 such as Random Access Memory (RAM) and/or Read Only Memory (ROM), Analog-to-Digital Converter (ADC) and a Digital-to-Analog Convertor (DAC), clocks, timers, counters and at least one processor (capable of implementing machine learning) connected with each other and to other components through communication bus channels. The memory element 112 is pre-stored with logics or instructions or programs or applications or modules/models and/or threshold values, safety threshold torque, safe threshold limit, which is/are accessed by the at least one processor as per the defined routines. The internal components of the controller 110 are not explained for being state of the art, and the same must not be understood in a limiting manner. The controller 110 is implementable in the form of System-in-Package (SiP) or System-on-Chip (SOC) or any other known types.

[0014] According to the present invention, a working of the controller 110 is explained. In a first working scenario or use case, the steering parameter, i.e. the steering angle or the rack position, is calculated by the controller 110 using the estimation model 114. Once calculated, the controller 110 uses the same for plausibility check with the steering angle/rack position calculated/determined based on the angle sensor 108, there by signals with lesser (B(D)) Automotive Safety Integrity Level (ASIL) integrity can be made to achieve better ASIL (D) integrity. Thus, the present invention enables to opt for a ASIL B (D) angle sensor 108 instead of a ASIL D sensor leading to cost reduction.

[0015] In a second working scenario or use case, the objective is to explore for a reliable and alternate solution for the cost sensitive physical steering angle sensors used in the steering wheel assembly 104, i.e. to develop a feasible system which serves as an alternative for the physical steering angle sensors. The objective is achieved by using the controller 110 to estimate the steering parameter, i.e. the steering angle/rack position using the estimation model 114, thereby improving accuracy and reliability across vehicle functions. Thus, in steering angle sensor units 102 where two angle sensors 108 are used, the present invention helps in replacing a second angle sensor with estimation model 114 thus reducing the per piece cost as one can opt for a single channel sensor instead of a dual channel sensor.

[0016] According to an embodiment of the present invention, the estimation model 114 replaces the physical angle sensors 108 completely after rigorous training on raw data.

[0017] According to the present invention, a method of training the estimation model 114 is disclosed. For example, once the steering angle sensor unit 102 is installed, the driver follows specific driving pattern. For example, forward driving - 40kmph. Forward driving with one left and one right turn > 15 kmph (both in single maneuver or two separate measurements). Reverse driving (straight) - safe speed. Reverse driving with one left and right turn > 15kmph (both in single maneuver or two separate measurements). Circular driving (parking lot scenario) - three turns at safe speed, etc. These are just for example and not to be understood in limiting manner.

[0018] Fig. 2 illustrates a method of operation of the steering angle sensor unit for the steering wheel assembly of the vehicle, according to the present invention. The method comprises plurality of steps of which a step 202 comprises determining, by the controller 110, the first steering parameter for the steering column 106 of the steering wheel assembly 104 using the angle sensor 108. The method is characterized by a step 204 which comprises estimating the second steering parameter, through the estimation model 114 stored in the memory element 112 of the controller 110. The at least one of the first steering parameter and the second steering parameter are used for steering parameter-dependent vehicle functions. The steering parameter is at least one of the steering angle and the rack position. Few examples of the vehicle functions are autonomous driving, active return of the steering, end stop of the steering, parking assist, and the like.

[0019] According to the method, the estimation model 114 comprises the physical model and the correction model. The physical model estimates the second steering parameter based on input parameters comprising yaw rate, lateral acceleration, and wheel velocity. The correction model comprises the Machine Learning/ Artificial Intelligence (ML/AI) based trained modules for correcting the second steering parameter estimated by the physical model.

[0020] According to the method, the second steering angle is correctable with at least one of the trained modules and real-time values of the angle sensor 108.

[0021] According to the method, the controller 110 is selected from any one of the first control unit and the combination of the first control unit and the second control unit. In the combination of the first control unit and the second control unit, the first control unit is interfaced with the angle sensor 108 to determine the first steering parameter and the second control unit is configured to estimate the second steering parameter.

[0022] According to the present invention, a method for estimating steering angle for the steering wheel assembly 104 of the vehicle 100 is disclosed. The method comprises plurality of steps of which a step 206 comprises of receiving signals for input parameters from respective means. The method is characterized by a step 208 which comprises estimating, by the controller 110, the steering parameter, i.e. the steering angle/rack position, through the estimation model 114 stored in the memory element 112 of the controller 110. The estimation model 114 comprises the physical model and the correction model. The input parameters are selected from yaw rate, lateral acceleration, and wheel velocity. The respective means corresponds to sensors to measure the yaw rate, lateral acceleration and wheel velocity or derived using models in dependence of other physical parameters of the vehicle 100.

[0023] The physical model estimates the second steering parameter based on input parameters comprising yaw rate, lateral acceleration and wheel velocity, and the correction model comprises the Machine Learning/ Artificial Intelligence (ML/AI) based trained modules to correct the second steering parameter estimated by the physical model. The training of the correction model is done during production/manufacturing stage before the vehicle 100 are rolled out for consumers. The training of the correction model is performed using the real-time value of the steering angle captured either from the angle sensor 108 or auxiliary sensor (another sensor to measure the steering angle) mounted solely for the training purposes.

[0024] According to the present invention, the present invention improves the ASIL integrity of the signals, in cost effective manner with reliability and scalability. The estimation model 114 which is ML/AI trained modules is used as an alternative for redundant second steering angle sensor. Due the present invention, the steering angle sensor unit 102 with lesser ASIL integrity can be used, i.e. ASIL B (D) is usable instead of ASIL D. Thus leading to cost saving. Further, thus steering angle sensor unit 102 with dual channel sensors is replaceable with single channel sensor. In simple words, a MI/AI based model as an alternative for steering angle sensor.

[0025] It should be understood that the embodiments explained in the description above are only illustrative and do not limit the scope of this invention. Many such embodiments and other modifications and changes in the embodiment explained in the description are envisaged. The scope of the invention is only limited by the scope of the claims.
, Claims:We claim:
1. A steering angle sensor unit (102) for a steering wheel assembly (104) of a vehicle (100), said steering angle sensor unit (102) comprises:
an angle sensor (108) mounted to a steering column (106) of said steering wheel assembly (104),
a controller (110) connected to said angle sensor (108) to determine a first steering parameter, characterized in that, said controller (110) configured to
estimate a second steering parameter, through an estimation model (114) stored in a memory element (112) of said controller (110), said second steering parameter is redundant of said first steering parameter, wherein at least one of said first steering parameter and said second steering parameter are used for steering parameter-dependent vehicle functions.

2. The steering angle sensor unit (102) as claimed in claim 1, wherein said estimation model (114) comprises a physical model and a correction model, wherein said physical model estimates said second steering parameter based on input parameters comprising yaw rate, lateral acceleration, and wheel velocity, and said correction model comprises a Machine Learning/ Artificial Intelligence (ML/AI) based trained modules to correct said second steering parameter estimated by said physical model.

3. The steering angle sensor unit (102) as claimed in claim 2, wherein said second steering parameter is correctable with trained modules and real-time values of said angle sensor (108).

4. The steering angle sensor unit (102) as claimed in claim 1, wherein said controller (110) is selected from any one of a first control unit and a combination of said first control unit and a second control unit, wherein in said combination of said first control unit and said second control unit, said first control unit is interfaced with said angle sensor (108) to determine said first steering parameter and said second control unit is configured to estimate said second steering parameter.

5. A controller (110) to estimate steering angle for a steering wheel assembly (104) of a vehicle (100), said controller (110) configured to:
receive signals for input parameters from respective means, characterized in that,
estimate a steering parameter, through an estimation model (114) stored in a memory element (112) of said controller (110), wherein said estimation model (114) comprises a physical model and a correction model, and said input parameters are selected from yaw rate, lateral acceleration, and wheel velocity.

6. A method of operation of a steering angle sensor unit (102) for a steering wheel assembly (104) of a vehicle (100), said method comprising the steps of:
determining a first steering parameter for a steering column (106) of said steering wheel assembly (104) using an angle sensor (108), characterized by,
estimating a second steering parameter, through an estimation model (114) stored in a memory element (112) of a controller (110), said second steering parameter is redundant of said first steering parameter, wherein at least one of said fist steering parameter and said second steering parameter are used for steering parameter-dependent vehicle functions.

7. The method as claimed in claim 6, wherein said estimation model (114) comprises a physical model and a correction model, wherein said physical model estimates said second steering parameter based on input parameters comprising yaw rate, lateral acceleration, and wheel velocity, and said correction model comprises a Machine Learning/ Artificial Intelligence (ML/AI) based trained modules for correcting said second steering parameter estimated by said physical model.

8. The method as claimed in claim 7, wherein said second steering parameter is correctable with at least one of said trained modules and real-time values of said angle sensor (108).

9. The method as claimed in claim 6 is executed by said controller (110) which is selected from any one of a first control unit and a combination of said first control unit and a second control unit, wherein in said combination of said first control unit and said second control unit, said first control unit is interfaced with said angle sensor (108) to determine said first steering parameter and said second control unit is configured to estimate said second steering parameter.

10. A method for estimating steering angle for a steering wheel assembly (104) of a vehicle (100), said method comprising the steps of:
receiving signals for input parameters from respective means, characterized in that,
estimating a steering parameter, through an estimation model (114) stored in a memory element (112) of said controller (110), wherein said estimation model (114) comprises a physical model and a correction model, and said input parameters are selected from yaw rate, lateral acceleration, and wheel velocity.