Online Exploratory Coverage Path Planning of Incremental SLAM for Autonomous Vehicles

This article aims to solve the exploratory coverage path planning (ECPP) and loop closure detection (LCD) problems of incremental simultaneous localization and mapping (iSLAM) for autonomous vehicles in an obstacle environment. An online ECPP method is proposed to enhance the overall performance of the planned trajectory while accommodating dynamic and incremental obstacle constraints. A novel probabilistic roadmaps trajectory homotopy (PRMTH) approach has been developed to navigate potential obstacles in environments with limited perception. An optimal multilayer map iSLAM method is proposed to update the map and improve the adaptability of obstacle avoidance for autonomous vehicles. Based on the ECPP method, ECPP and LCD mapping algorithms are developed considering the mapping coverage rate, trajectory length, and turning radius constraints. The final planning trajectory of the LCD is smoothed using the Dubins algorithm. A grid-based iSLAM method is proposed to address the challenges of map updating for autonomous vehicles operating in obstacle-rich environments. The results of the simulations and experiments show that the proposed ECPP and LCD strategies can provide an optimal solution to incremental simultaneous localization and mapping (SLAM) for autonomous vehicles.