We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
A Voronoi path planning extracted from improved skeleton for dynamic environments.
- Authors
Jiang, Lin; Li, Jun; Hu, Yuxin; Pan, Feng; Zhu, Jianyang; Lei, Bin; Lin, Rui
- Abstract
Aiming at the problems that the robot being in the process of navigation cannot meet the requirements of real-time and accuracy at the same time, moreover is too close to obstacles and lacks the initiative to avoid obstacles, a Voronoi diagram algorithm for improved skeleton extraction suitable for dynamic environment is proposed. On the one hand, firstly the grid map is preprocessed by binarization, corrosion and expansion, so the reduced skeleton map suitable for navigation is obtained, then the reduced skeleton map is extracted for searching the global path, finally the improved cubic spline smoothing algorithm is used to optimize the global path each planned, thus overcoming the defects of bloated and tortuous in the path obtaining by original Voronoi diagram algorithm. On the other hand, the position information of all obstacles is obtained by a single scan lidar. Firstly, segmenting and linearly fitting all laser point clouds to remove the known obstacles in the map. Then to mark new possible dynamic obstacles with circles of appropriate size. Secondly detecting dynamic obstacles by the alteration of their center coordinates, moreover, solving their motion equations. Finally expanding the cost map along the speed direction of dynamic obstacles and combining DWA dynamic window method to realize dynamic obstacle avoidance. Compared with the original DWA algorithm, it can predict the motion state of dynamic obstacles in advance, which improves the safety of the robot in the dynamic environment. Moreover, the effectiveness of the algorithm is verified by many simulation experiments and real environment experiments.
- Subjects
VORONOI polygons; GRIDS (Cartography); EQUATIONS of motion; NAUTICAL charts; POINT cloud
- Publication
Journal of Mechanical Science & Technology, 2023, Vol 37, Issue 4, p2019
- ISSN
1738-494X
- Publication type
Article
- DOI
10.1007/s12206-023-0338-4