Efficient trajectory planning method in multilayer structured environment
By extracting the cross-border planes in 3D point clouds and optimizing the motion state trajectory, the problem of trajectory planning in complex multi-layer environments is solved, and efficient and safe robot navigation is achieved.
CN119690081AActive Publication Date: 2025-03-25HUZHOU INST OF ZHEJIANG UNIV
4 Cites 0 Cited by
Patent Information
- Application Number
- CN202411849504.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-16
- Publication Date
- 2025-03-25
- Estimated Expiration
- 2044-12-16
AI Technical Summary
Technical Problem
The prior art is difficult to generate efficient, smooth and secure robotic trajectories in complex multi-layer structured environments, especially when climbing and down stairs.
Method used
By extracting crossable planes from 3D point clouds, trajectory optimization is built based on motion state trajectories, and special constraints are taken into account when crossing multi-layer planes to maximize the maneuverability of the robot.
Benefits of technology
Generating safe and kinematic appropriate trajectories in complex multi-story building spaces is achieved, and the navigation efficiency and safety of robots in multi-story environments is improved.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention provides an efficient trajectory planning method in a multi-layer structured environment, which is used for autonomous navigation of a ground robot in a large complex multi-layer building space, simplifying the problem into navigation between complex intersecting planes, extracting a traversal plane from a 3D point cloud in advance, constructing trajectory optimization based on a motion state trajectory, and solving the problem of trajectory optimization. And special constraints are fully considered when the robot passes through multiple layers of planes, so that the maneuverability of the robot is maximized.
Need to check novelty before this filing date? Find Prior Art