Trajectory generation and optimization method and system based on spatiotemporal probability road tree

The initial trajectory is generated by a method based on the spatiotemporal probability road tree and combined with the time elastic band and model prediction control for optimization, the problem of difficult to obtain the initial solution of trajectory planning in complex dynamic environments is solved, and efficient and safe trajectory planning and obstacle interaction are achieved.

CN114625141BActive Publication Date: 2025-03-21XI AN JIAOTONG UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202210260195.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-16
Publication Date
2025-03-21
Estimated Expiration
2042-03-16

AI Technical Summary

Technical Problem

In complex dynamic environments, it is difficult for the prior art to effectively obtain high-quality initial feasible solutions, resulting in limited trajectory planning optimization performance, especially when interacting with dynamic and static obstacles in dynamic environments, which are difficult to achieve safe and optimized trajectories.

Method used

The trajectory generation and optimization method based on the space-time probability road tree is adopted, and the space-time road tree is constructed through discretization and random sampling of the space-time domain to generate the initial trajectory, and optimized with the time elastic band method and the model prediction and control method to achieve unified processing of dynamic and static obstacles.

Benefits of technology

Improve the optimization performance and efficiency of trajectory planning, and can generate approximately optimal trajectories in complex dynamic environments, ensure safe interaction between robots and obstacles, and have predictive capabilities to respond to future behaviors of obstacles.

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

Abstract

The invention discloses a trajectory generation and optimization method and system based on a spatiotemporal probability road tree. The method comprises the following steps: determining a reference state sequence of a robot and a free area of ​​each future frame; performing point random sampling in the free area of ​​each frame, and if the sampling points of the current frame and the sampling points of the previous frame meet the condition of building an edge, building an edge, and adding a loss value to the sampling points of the current frame according to the reference state sequence, until all the sampling points of the last frame are judged, and building a spatiotemporal probability road tree and the loss values ​​of the sampling points in the spatiotemporal probability road tree; searching for a path with the minimum loss, i.e., an initial trajectory, based on the loss values ​​of the sampling points in the spatiotemporal probability road tree; determining the boundary limit of each frame state on the initial trajectory, and optimizing the initial trajectory according to the reference state sequence and the objective function; until the robot kinematic constraint and the obstacle constraint are met or the iteration upper limit is reached to obtain the optimized trajectory.
Need to check novelty before this filing date? Find Prior Art