Multi-AGV's Temporal Memory-Based RRT Exploration in Unknown Environment

With the increasing need for multi-robot exploring unknown region in a challenging environment, efficient collaborative exploration strategies are needed achieving such feat. A frontier-based Rapidly-Exploring Random Tree (RRT) can be deployed to explore an environment. However, its' greedy behavior causes multiple robots with highest revenue, which leads massive overlapping process. To address this issue, we present temporal memory-based RRT (TM-RRT) strategy perform robust It computes adaptive duration each frontier assigned and calculates frontier's revenue based on relative position of robot. In addition, robot is equipped memory consisting share among fleets prevent repeating assignment same frontier. Through both simulation actual deployment, have shown robustness TM-RRT by completing $25.0\,\rm{m}\times \text{54.0}\,\rm{m}$ ( notation="LaTeX">$\text{1350.0}\,\rm{m}^{2}$ ) area, while conventional falls short.