Obstacle Avoidance Algorithm for Multi-Intelligent Body Formation System Based on Artificial Potential Field Approach

This paper proposes a cooperative strategy integrating distributed mean-shift and the Artificial Potential Field (APF) approach to solve formation shaping with obstacle avoidance in multi-agent systems. Mean-Shift algorithm: Meanshift algorithm: The algorithm drives agents toward target regions by f...

Full description

Saved in:
Bibliographic Details
Published in:2025 10th International Conference on Intelligent Computing and Signal Processing (ICSP) pp. 816 - 819
Main Author: Zhang, Shunyao
Format: Conference Proceeding
Language:English
Published: IEEE 16.05.2025
Subjects:
Artificial Potential Field Approach
Collision avoidance
Heuristic algorithms
Mean-Shift
Multi-agent systems
Navigation
Obstacle avoidance
Robot sensing systems
Robustness
Thermal stability
Three-dimensional displays
Topology
Trajectory optimization
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a cooperative strategy integrating distributed mean-shift and the Artificial Potential Field (APF) approach to solve formation shaping with obstacle avoidance in multi-agent systems. Mean-Shift algorithm: Meanshift algorithm: The algorithm drives agents toward target regions by following local grayscale gradients. A velocity compensation term is introduced to adjust movement direction, while density repulsion within sensing ranges balances agent distribution, thus addressing uneven deployment issues. Artificial Potential Field Approach: Obstacles generate repulsive potential fields. When a robot approaches an obstacle, the APF function computes a repulsive force inversely proportional to their separation distance, steering the robot away to ensure collision-free navigation. Simulation results verify that the proposed method successfully enables multi-agent systems to form complex formations and dynamically perform obstacle avoidance, displacement, and reorganization in cluttered environments.
DOI:10.1109/ICSP65755.2025.11086828