Research on Motion Planning Based on Flocking Control and Reinforcement Learning for Multi-Robot Systems

Abstract

Robots have poor adaptive ability in terms of formation control and obstacle avoidance control in unknown complex environments. To address this problem, in this paper, we propose a new motion planning method based on flocking control and reinforcement learning. It uses flocking control to implement a multi-robot orderly motion. To avoid the trap of potential fields faced during flocking control, the flocking control is optimized, and the strategy of wall-following behavior control is designed. In this paper, reinforcement learning is adopted to implement the robotic behavioral decision and to enhance the analytical and predictive abilities of the robot during motion planning in an unknown environment. A visual simulation platform is developed in this paper, on which researchers can test algorithms for multi-robot motion control, such as obstacle avoidance control, formation control, path planning and reinforcement learning strategy. As shown by the simulation experiments, the motion planning method presented in this paper can enhance the abilities of multi-robot systems to self-learn and self-adapt under a fully unknown environment with complex obstacles.

Fuzzy-Based Distributed Behavioral Control With Wall-Following Strategy for Swarm Navigation in Arbitrary-Shaped Environments

“…The same pattern could be observed when the robot swarm encountered obstacles 2 and which show the adaptability of our approach to different environments. In addition, while researchers in [14]- [18] allowed swarm robots to communicate to synchronize controller or behavior switching, Fig. 7 shows that multi synthesized controllers or states coexisted and independent at each time step.…”

“…In the second category, instead of the obstacle avoidance behaviours, the wall-following strategy is used to overcome the obstacle by following its boundary. Most research focusing on wall-following strategy was performed on a single robot while very few ones are found in multi-robot systems with challenges in cooperation and interaction between robots [14]- [18]. The swarm configuration in these wall- following strategies can be categorized into a one-chain (straight line) [16]- [18], a formation [15], a flexible configuration as the rolling motion in the vortex pattern [14].…”

“…Most research focusing on wall-following strategy was performed on a single robot while very few ones are found in multi-robot systems with challenges in cooperation and interaction between robots [14]- [18]. The swarm configuration in these wall- following strategies can be categorized into a one-chain (straight line) [16]- [18], a formation [15], a flexible configuration as the rolling motion in the vortex pattern [14]. Specifically, the flocking behaviour changes to the rolling motion behaviour in the vortex pattern to make the robot swarm to follow the boundary of the non-convex obstacle in u-shape [14].…”

“…The control strategy uses only local sensing the relative distance and bearing measured by infrared sensors and tested with two robots. In [18], the robot swarm moves from a source to destination in a formation (such as diamond) with leader-follower flocking control. If the formation encounters the obstacle that prevents the robots from maintaining their formation, the formation is transformed into a line formation for wall-following strategy.…”

“…The reinforcement learning is adopted to implement the robot behavioral selection from a set of behaviours including goal approaching, obstacle avoidance, collision avoidance, wall-following movement through predicting abilities of the robot during motion. In these few existing studies, most studies is not distributed control requiring communication for configuration transition as well as configuration maintenance [14]- [16], [18], excepts [17] that only considers the wall-following control without configuration transition. Table 1 summarizes the differences between our proposed approach and the previously reviewed works.…”