Fpga Based Multi-Robot Collision Avoidance System

Abstract

This paper presents a Multi-robot collision avoidance system suitable for FPGA implementation. The proposed scheme attempts a novel method of Reciprocal Collision Avoidance(RCA) along with Accelerating Velocity Obstacles(AVO). Parallel architecture is used for developing the collision analysis of robots with different targets. Each of these robots moves and reaches its target, avoiding collisions on its path. Simulations for the digital design hardware have been carried out in Vivado software and the robotic movements have been demonstrated using V-rep software. The proposed scheme would result in reduced cost and power dissipation while deploying in FPGAs. The details of the collision detection algorithm and the working of different modules used in the scheme along with test results are presented in the paper.

Plugin Framework-Based Neuro-Symbolic Grounded Task Planning for Multi-Agent System

“…Various studies on algorithms are being conducted on simultaneous localization and mapping (SLAM) [ 15 , 16 , 17 ], collision avoidance [ 18 , 19 , 20 ], and formation [ 21 , 22 ] for multi-agent systems. Jang et al [ 23 ] introduced a collaborative monocular SLAM using the rendezvous generated as multiple robots execute tasks.…”