Iterative Feedback Motion Planning

Description:

Reference #: 01394

The University of South Carolina is offering licensing opportunities for Iterative Feedback Motion Planning

Background:

In an autonomous driving system (ADS), the motion planning module is responsible for generating a motion trajectory for the motion controller. The motion controller module attempts to drive the vehicle while following the motion trajectory as closely as possible. Although a lot of advanced control methods such as using a linear quadratic regulator or model predictive controller improve the tracking performance, their effectiveness can be improved upon. An invention method is proposed here to establish a communication from the motion controller to the motion planner to further improve the tracking performance.

Invention Description:

This technology proposes iterative trajectory optimization (ITO) to improve the motion controller’s tracking performance and increase the physical and operational feasibility of motion planning trajectory. ITO uses constant communication between the motion controller and the motion planner. This system diverges from the traditional motion controller by using two sub-modules: the iterative motion simulator and the motion control operator. In the iterative motion simulator, the vehicle trajectory response is simulated while updating a trajectory offset with each iteration, correcting the vehicle response to be closer to the reference trajectory. After the iterative trajectory adjustment finishes, the simulated tracking error and trajectory offset are sent back to motion planner. The motion planner will first evaluate the simulated trajectory to see the response effectiveness, and then the trajectory with the offset is finally sent to the motion control operator to calculate the vehicle control maneuver. Comparing this with the traditional motion planning system, the proposed ITO approach can guarantee the trajectory’s physical and operational feasibility and allow the motion controller to have improved tracking performance and have a predictive evaluation on how the vehicle will respond to inputs. The simulation results demonstrate the effectiveness of the proposed method.

Potential Applications:

This technology advancement has specific uses in automotive engineering in autonomous cars. Any automotive company with interest in developing a successful autonomous car could benefit from this technology, including major automotive OEMs, Tesla, Uber, and Lyft. Optimizing the data processing capability of autonomous driving systems is a critical component to allowing the commercialization of these systems. 

Advantages and Benefits:

This technology allows for autonomous cars to be able to more closely follow the proposed path by the motion planner. This enables more precise movement and less error in trajectory as the car is planning its own path. 

 

Patent Information:
Title App Type Country Serial No. Patent No. File Date Issued Date Expire Date Patent Status
Iterative Feedback Motion Planning Utility United States 16/804,094   2/28/2020     Published
For Information, Contact:
Technology Commercialization
University of South Carolina
technology@sc.edu
Inventors:
Bin Zhang
Zhichao Liu
Kai Zhang
Keywords:
feedback planning
Iterative learning
motion control
motion planning
© 2024. All Rights Reserved. Powered by Inteum