A Leader Follower Algorithm Based Path Navigation Approach for Multiple Autonomous Vehicles

Abstract

Multi-agent path navigation in unknown environments is a challenging task where an effective device-to-device communication-based algorithm needs to be designed with minimum error. A leader-follower algorithm has been proposed in this literature for multi-agent communication. The proposed algorithm has improved the percentage deviation in travel time and the percentage deviation in path length. The proposed algorithm is designed in two parts. One part is for leader path navigation, and the other is used to update the followers. The obstacle avoidance mechanism is used in the first part of the algorithm. The computation speed of the algorithm is mainly dependent on the guidance that the leader provides to the follower agent. Path update from the leader side to the follower side is the most important task in this proposed method. One leader and two follower agents were considered in simulation environments to establish the feasibility of the algorithm. In the simulation, each follower agent was made to commute with a few seconds' time gap after the leader had started. After several simulations, the proposed approach has a 2.82% average deviation in travel time and a 31% average deviation in path length. The average success rate of the proposed technique in different environments was found equal to 66%. Studies show that the accuracy of this technique is more than 54% on average.