Reinforcement learning based adaptive planner parameter tuning method and apparatus

By employing a hierarchical architecture of low-frequency parameter tuning, mid-frequency planning, and high-frequency control, combined with a reinforcement learning error compensator, the limitation of the control layer in adaptive planner parameter tuning is solved, enabling efficient and accurate navigation of the robot in complex environments.

CN119644733BActive Publication Date: 2025-11-28ZHEJIANG UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202411760008.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-12-03
Publication Date
2025-11-28
Estimated Expiration
2044-12-03

AI Technical Summary

Technical Problem

Existing adaptive planner parameter tuning methods focus only on optimization at the tuning level in trajectory planning, ignoring the constraints of the control level. This leads to tracking errors and noise problems, making it unable to effectively cope with dynamic changes in complex environments.

Method used

A hierarchical architecture of low-frequency parameter tuning, mid-frequency planning, and high-frequency control is adopted. Combined with a reinforcement learning error compensator, the local planner and controller parameters are optimized through an alternating training framework, thereby reducing tracking errors and improving system robustness.

Benefits of technology

It significantly reduces trajectory tracking errors, improves the stability and efficiency of robot navigation, enhances obstacle avoidance capabilities, and enables more accurate and reliable navigation task execution.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a kind of based on reinforcement learning's adaptive planner parameter tuning method and device, method includes: at low frequency control frequency, in parameter adjustment network, according to the input environmental data, using reinforcement learning algorithm to adaptively adjust the parameters of local planner;At medium frequency control frequency, in local planner, according to the adjusted parameter, the planning trajectory of robot travel is generated and the feedforward speed is generated according to the planning trajectory;At high frequency control frequency, in controller, according to the feedforward speed and the feedback speed calculated by error compensator, generate final control instruction;Parameter adjustment network and controller are alternately trained and tested iteratively, and according to the test result, the parameters of local planner and the final control instruction generated by controller are continuously optimized.The application can improve system robustness and navigation efficiency, effectively reduce tracking error and enhance obstacle avoidance ability, can complete more accurate, efficient and reliable robot navigation task in complex environment.
Need to check novelty before this filing date? Find Prior Art