A Self-organizing Multi-agent Cooperative Robotic System: An Application of Cohort Intelligence Algorithm

Abstract

This paper presents an application of the emerging Cohort Intelligence (CI) algorithm in the domain of swarm robotics. The application could be relevant to search and rescue in alien territory as well as establishment. The robots are considered as candidates in the CI algorithm. An exponential probability approach is proposed by which every candidate/robot decide to follow one another. In this approach, the probability of following the worse candidate decreases and the probability stake of the better candidate increases. This makes the robots more biased to follow better candidates. This helps to reduce the randomness in the system. The approach was applied and validated by solving path planning and obstacle avoidance for application of a swarm of robots in a static and unknown environment. The cases such as No Obstacle Case (NOC), Rectangular Obstacle Case (ROC), Multiple Rectangular Obstacles Case (MROC) and Cluttered Polygonal Obstacles Case (CPOC) were solved. The results obtained were better in terms of computational time and function evaluations as compared to the linear probability distributions approach. The limitations of the approach solving the obstacle avoidance for swarm of robots are also discussed.