US 12,447,957 B2
Automotive vehicle control circuit
Xiao-Dong Sun, Solihull (GB)
Assigned to ZF Automotive UK Limited, Solihull (GB)
Appl. No. 18/270,039
Filed by ZF Automotive UK Limited, Solihull (GB)
PCT Filed Dec. 30, 2021, PCT No. PCT/GB2021/053446
§ 371(c)(1), (2) Date Jun. 28, 2023,
PCT Pub. No. WO2022/144550, PCT Pub. Date Jul. 7, 2022.
Claims priority of application No. 2020838 (GB), filed on Dec. 31, 2020.
Prior Publication US 2024/0075924 A1, Mar. 7, 2024
Int. Cl. B60W 30/12 (2020.01)
CPC B60W 30/12 (2013.01) [B60W 2510/20 (2013.01); B60W 2552/53 (2020.02); B60W 2710/20 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A lane keep assist (LKA) system for an automotive Vehicle which includes an electric power steering assembly that is responsive to an output of a control system, the motor applying a torque to a part of a steering gear to steer the vehicle along a highway, the lane keep assist system assisting a driver in keeping the vehicle in a lane of a highway, in which the control system comprises:
a PID Controller configured to receive at an input a target lane position for the closed-loop control system and provides as an output a control signal for a motor of the electric power steering assembly, the PID controller being arranged in a closed loop configuration with the motor configured to minimise an error value indicative of the difference between the target lane position and the actual lane position of the vehicle, and
a neural network including an input layer of neurons, at least one hidden layer of neurons, and an output layer comprising at least one output neuron,
in which the neural network comprises a feedforward neural network that receives at the input layer of input neurons the target lane position, the control signal output from the PID controller and the error value,
and in which the neural network is configured to determine the P gain, I gain and D gain terms used by the PID controller, in which the neural network determines the gain values as respective nodal values within a hidden layer of the neural network via gradient-descent backpropagation learning, wherein a respective gain term corresponds to a different neuron within the neural network,
and further in which the neural network receives as a feedforward term one or more of the vehicle speed and the curvature of the lane the vehicle is to be kept in.