Trajectory planning system for a vehicle and vehicle
A trajectory planning system for a vehicle includes a sensor system for measuring an actual curvature on the basis of a respective current yaw rate, a memory containing a target trajectory with target curvatures, wherein the trajectory planning system determines target yaw rates from the target curvature, and a steering system that uses steering variables to obtain target curvatures in actual curvatures. The system determines a respective first actual derivative of the measured respective current actual yaw rate over time and a respective first target derivative of the respective current target yaw rate over time. A correlator determines a respective current delay on the basis of the respective first actual derivatives and the respective first target derivatives in a current yaw rate segment, and a parameter estimator recursively estimates the delay between the target yaw rate and the actual yaw rate on the basis of respective current delay inputs.
1 . A trajectory planning system for a vehicle comprising:
a sensor system for measuring an actual curvature on a basis of a respective current actual yaw rate;
a memory containing a target trajectory with target curvatures;
a steering system with steering variables configured to be applied to an actual curvature to obtain the target curvatures using the steering variables; and
at least one processing device configured to:
determine target yaw rates from the target curvature in the target trajectory;
determine a respective first actual derivative of the measured respective current actual yaw rate over time and a respective first target derivative of the respective current target yaw rate over time;
determine a respective current delay on a basis of the respective first actual derivative and the respective first target derivative in a current yaw rate segment;
recursively estimate an estimated delay between the target yaw rate and the actual yaw rate on a basis of respective current delay inputs; and
determine a time period on a basis of at least the estimated delay in relation to the actual curvature, in order to compensate for a delay while obtaining the target curvature by applying the steering variables to the actual curvature.
2 . The trajectory planning system according to claim 1 , comprising:
a control system configured to set the steering variables for the steering system, wherein the control system is configured to set the steering variables for the desired vehicle performance in advance of the time period.
3 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
determine a second actual derivative of the measured respective current actual yaw rate over time and a second target derivative of the respective current target yaw rate over time; and
determine the respective current yaw rate segment, wherein the respective current yaw rate segment has a maximum length in which the second actual derivative and the second target derivative each exceed a predefined positive and negative threshold value at least once.
4 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
recursively estimate the estimated delay on a basis of respective current delay and vehicle parameters, wherein the vehicle parameters comprise at least the speed of the vehicle.
5 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
recursively estimate the estimated delay on a basis of respective current delays and vehicle parameters, as well as environmental input data, wherein the vehicle parameters comprise at least the speed of the vehicle.
6 . The trajectory planning system according to claim 5 , wherein the environmental data comprise at least a nature of a road surface.
7 . The trajectory planning system according to claim 5 , wherein the environmental data comprise at least a detected frictional coefficient.
8 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
provide a recursive least squares estimator.
9 . The trajectory planning system according to claim 8 , wherein the recursive least squares estimator contains a forgetting factor for deleting outdated results.
10 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
provide an artificial neural network.
11 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
optimize internal parameters for estimating the estimated delay; and
use the optimized internal parameters for future estimations of the estimated delay
once the optimized internal parameters reach a certain level of precision.
12 . The trajectory planning system (according to claim 1 , wherein the at least one processing device is configured to:
determine the time period in realtime.
13 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
determine the current delay by cross-correlation of the respective first actual
derivatives and the respective first target derivatives in the current yaw rate segment.
14 . The trajectory planning system according to claim 1 , wherein the at least one processing device is configured to:
determine a new estimated delay in response to a newly determined current delay.
15 . A vehicle comprising:
the trajectory planning system according to claim 1 ; and
a receiver configured to receive the target curvature at predefined locations from one or more leading vehicles.
16 . A vehicle comprising:
the trajectory planning system according to claim 1 ; and
a trajectory generating system configured to generate the target trajectory on a basis of at least navigation data and environmental data.
17 . A method for trajectory planning comprising:
measuring an actual curvature on a basis of a respective current actual yaw rate;
providing a target trajectory with a target curvature, and determining target yaw rates from the target curvature in the target trajectory;
providing a steering system with steering variables for implementing the target curvatures on the basis of the steering variables;
determining a respective first actual derivative for the measured respective current actual yaw rate and a respective first target derivative for the respective current target yaw rate over time;
determining a respective current delay on a basis of the respective first actual derivative and the respective first target derivative in a current yaw rate segment with a correlator;
recursively estimating an estimated delay between the target yaw rate and the actual yaw rate on a basis of the respective current delay input into a parameter estimator; and
determining a time period on a basis of at least the estimated delay in relation to the actual curvature by a preview estimator in order to compensate for a delay in obtaining the target curvature by applying the steering variables to the actual curvature.