Autonomous automotive vehicle lateral motion control system

Abstract

An autonomous vehicle (101) lateral motion control system comprising: a trajectory planner (102); and a trajectory controller (104) generating a steering control signal for controlling the lateral motion of the vehicle. The trajectory controller comprises a control circuit (104a) comprising an integrated neural network adaptive PID controller (124) which comprises: a PID controller (116) which receives a target lane position and is arranged to minimize an error value indicative of the difference between the actual trajectory of the vehicle and the target lane position dependent input signal, the PID controller being further arranged to receive a target course angle and is arranged to minimize an error value indicative of the difference between the actual trajectory of the vehicle and the target course angle dependent input signal; and a neural network (118) configured to determine the P gain, I gain and the D gain terms used by the PID controller. The integrated neural network adaptive PID controller is configured to use the P and I gain terms provided by the neural network to control the lane position of the vehicle, and the D gain term provided by the neural network to control the course angle of the vehicle.

Claims

1. An autonomous vehicle lateral motion control system, comprising:a trajectory planner configured to generate a target trajectory for the vehicle; anda trajectory controller configured to receive the target trajectory from the trajectory planner and generate as an output a steering control signal for controlling the lateral motion of the vehicle;wherein the trajectory controller comprises a control circuit, the control circuit comprising an integrated neural network adaptive PID controller which comprises:a PID controller which receives an input signal dependent on a target lane position parameter of the target trajectory and is arranged to minimise an error value indicative of the difference between the actual trajectory of the vehicle and the target lane position parameter dependent input signal, the PID controller being further arranged to receive an input signal dependent on a target course angle parameter of the target trajectory and is arranged to minimise an error value indicative of the difference between the actual trajectory of the vehicle and the target course angle parameter dependent input signal; anda neural network configured to determine the P gain, I gain and the D gain terms used by the PID controller,wherein the integrated neural network adaptive PID controller is configured to use the P and I gain terms provided by the neural network to control the lane position of the vehicle, and the D gain term provided by the neural network to control the course angle of the vehicle, whereby the P, I and D gain terms used by the PID controller are provided by the same neural network.

2. A motion control system according to claim 1, wherein the trajectory controller further comprises a curvature controller configured to perform one or more of P, I, and D control of the curvature of the trajectory of the vehicle.

3. A motion control system according to claim 2, wherein the curvature controller uses pre-determined and / or vehicle motion variable dependent P gain, I gain and / or D gain terms which do not change during use of the motion control system.

4. A motion control system according to claim 2, wherein the curvature controller is a neural network adaptive controller having a neural network configured to provide P gain, I gain and / or D gain terms.

5. A motion control system according to claim 2 or claim 3 or claim 4, wherein: the curvature controller is configured to receive as an input a target curvature parameter and generate as an output a curvature control signal;the trajectory controller is configured to determine the sum of the curvature control signal and the output of the integrated neural network adaptive PID controller to provide the steering control signal output for the trajectory controller.

6. A motion control system according to any preceding claim, wherein the neural network comprises an input layer of neurons, at least one hidden layer of neurons, and an output layer comprising at least one output neuron, wherein the neural network comprises a feedforward neural network that receives at the input layer of input neurons the error value associated the input trajectory parameter and the control signal output by the integrated neural network adaptive PID controller, and outputs the gain values at respective nodal values within the hidden layer of the neural network.

7. A motion control system according to claim 6, wherein the neural network is further configured to receive as a feedforward term at least one environmental variable.

8. A motion control system according to claim 7, wherein the environmental variable comprises at least one of the following commonly found in a vehicle motion control system:-the speed of the vehicle for most of the vehicle motion control,- road / traffic measurements such as the curvature, heading and speed limit etc.- vehicle acceleration / reaction force- the motor rotation speed / angle for the current / torque control- steering torque or angle-vehicle turn signal condition, e.g. turn indicated or turn not indicated9. A motion control system according to any of claims 6 to 8, wherein during use of the control system the weights of the neural network are updated by a gradientdescent backpropagation scheme each time a new set of input values is supplied to theneural network, and the updated weights combined with the input values are used to update the neurons.

10. An autonomous automotive vehicle comprising the motion control system of any 5 preceding claim.

Images