Multi-Robot Directed Coverage Path Planning in Row-based Environments
Tingjun Lei, Pradeep Chintam, Chaomin Luo, Shahram Rahimi
Abstract
Multiple autonomous robots are deployed to fulfill tasks collaboratively in real-world applications with row-based settings as found in precision agriculture, warehouses, factory inspections, and wind farms. One batch of robots are assigned to explore, search and localize objects in large-scale row-based environments, while the other batch of robots move directly to the detected targets to retrieve the objects. In this paper, a multi-robot collaborative navigation framework with two different batches of robots is proposed to explore the environment and achieve the obtained targets, respectively. The first batch of robots act as detection robots, which are driven by a proposed informative-based directed coverage path planning (DCPP) through a multi-robot minimum spanning tree algorithm. It refines and optimizes the coverage path based on the information gained from the environment. The second batch of robot reaches the multiple targets by guidance from a hub-based multi-target routing (HMTR) scheme, which is applicable to row-based environments. The feasibility and effectiveness of the proposed methods are validated by simulation and comparison studies.