{"type":"ScholarlyArticle","title":"Deep Reinforcement Learning for Multi-contact Motion Planning of Hexapod Robots","authors":[{"name":"Huiqiao Fu","slug":"huiqiao-fu","given":"Huiqiao","family":"Fu","affiliation":""},{"name":"Kaiqiang Tang","slug":"kaiqiang-tang","given":"Kaiqiang","family":"Tang","affiliation":""},{"name":"Peng Li","slug":"peng-li","given":"Peng","family":"Li","affiliation":""},{"name":"Wenqi Zhang","slug":"wenqi-zhang","given":"Wenqi","family":"Zhang","affiliation":""},{"name":"Xinpeng Wang","slug":"xinpeng-wang","given":"Xinpeng","family":"Wang","affiliation":""},{"name":"Guizhou Deng","slug":"guizhou-deng","given":"Guizhou","family":"Deng","affiliation":""},{"name":"Tao Wang","slug":"tao-wang","given":"Tao","family":"Wang","affiliation":""},{"name":"Chunlin Chen","slug":"chunlin-chen","given":"Chunlin","family":"Chen","affiliation":""}],"abstract":"Legged locomotion in a complex environment requires careful planning of the footholds of legged robots. In this paper, a novel Deep Reinforcement Learning (DRL) method is proposed to implement multi-contact motion planning for hexapod robots moving on uneven plum-blossom piles. First, the motion of hexapod robots is formulated as a Markov Decision Process (MDP) with a specified reward function. Second, a transition feasibility model is proposed for hexapod robots, which describes the feasibility of the state transition under the condition of satisfying kinematics and dynamics, and in turn determines the rewards. Third, the footholds and Center-of-Mass (CoM) sequences are sampled from a diagonal Gaussian distribution and the sequences are optimized through learning the optimal policies using the designed DRL algorithm. Both of the simulation and experimental results on physical systems demonstrate the feasibility and efficiency of the proposed method. Videos are shown at https://videoviewpage.wixsite.com/mcrl.","doi":"10.24963/ijcai.2021/328","url":"https://ijcai.org/proceedings/2021/328","pdf":"https://ijcai.org/proceedings/2021/0328.pdf","markdown":"https://ijcai.org/proceedings/2021/328.md","bibtex":"https://ijcai.org/proceedings/2021/bibtex/328","volume":{"slug":"2021","title":"Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2021","url":"https://ijcai.org/proceedings/2021/","year":2021,"location":"Montreal","dates":"19-27 August 2021","editors":"Prof. Zhi-Hua Zhou","isbn":"978-0-9992411-9-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2021"},"section":"Machine Learning","track":"Main Track","number":328,"pages":{"first":2381,"last":2388},"keywords":["Machine Learning: Deep Reinforcement Learning","Robotics: Motion and Path Planning","Robotics: Learning in Robotics"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}