Autonomous Robotic Exploration
Autonomous Robotic Exploration of Unknown Campus Environment
Autonomous robotic exploration in an unknown campus environment is a big challenge for a mobile robot. Penetrable obstacles (e.g. railings, glass, and curbs) in the environment interfere with the computation of exploration gains, low obstacles (e.g. curbs and level lawns) affect the accuracy of traversable area analysis, frequent loop closures make frontier information difficult to maintain and exploration maps difficult to reuse. To address these problems, we propose a novel robotic exploration framework in an unknown campus environment. This framework consists of two parts: a compact 3D map representation and an efficient exploration strategy in alignment with an online SLAM procedure.
Reference paper:
Zezhou Sun, Banghe Wu, Chengzhong Xu, and Hui Kong, Concave-Hull Induced Graph-Gain for Fast and Robust Robotic Exploration, IEEE Robotics and Automation Letters (RA-L), 2023,
Zezhou Sun, Banghe Wu, Chengzhong Xu, Hui Kong, Ada-Detector: Adaptive Frontier Detector for Rapid Exploration, IEEE International Conference on Robotics and Automation (ICRA),https://arxiv.org/abs/2204.06237, 2022