Multi-robot cooperation and path planning for stick transporting using improved Q-learning and democratic robotics PSO

摘要

The highlight of this paper is to propose an innovative method for solving the problem of path planning and synchronization of robot pairs in a known static and complex environment. The problem addresses the stick transportation by robot pairs from the predefined starting position to the pre-assumed goal position. This paper describes the method of computing collision-free paths for multiple numbers of paired robots called twins using a hybrid algorithm with balanced local exploitation and global exploration. The hybrid algorithm is designed by combining the individual benefits of the improved Q-Learning (IQ) and democratic robotics particle swarm optimization (DRPSO). The main intention of IQ is to produce the best position with collision avoidance by executing the best state-action pair. DRPSO algorithm computes the optimal trajectory for the twin robot using the distributed democratic approach and enhances the intensification capability. The twin robots are synchro-nized to the number of steps, turns, the direction of movement, and distance between both of them in the proposed approach. Computer simulation results reveal the efficacy of the proposed algorithm in both paths travelled and runtime. The experimental and VREP simulation results illustrate that the proposed algorithm offers better synchronization, cooperation, and performance in terms of several parameters such as path length, path deviation, run time, and path smoothness as a replica of the number of turns made. The comparative study with artificial bee colony optimization (ABCO) and imperialist competitive firefly algorithm (ICFA) witnesses the superiority of the hybrid IQ-DRPSO algorithm. Experiments that have been conducted through Arduino bots in a real environment prove the efficiency of the proposed approach. Results obtained from simulation and experi-ment reveal the supremacy of the proposed algorithm for synchronization, cooperation, and path traveled.