Abstract
AbstractAdopting the concept of the tactile wheel, which considers the interaction between the wheel and the ground, this paper simulates reinforcement learning to show the usefulness of tactile sensing for autonomous wheeled robots on irregular terrain and to clarify the characteristics of the information to be acquired. A wheeled robot model with a wheel-on-leg structure is created and tested on two types of irregular terrain. The tactile information from each wheel is used as part of the reinforcement learning state. The average return and sample efficiency respectively increase by factors of 1.18 and 2.21 on uneven terrain. On fractal terrain, they increase by factors of 1.31 and 2.51 times, respectively, confirming the usefulness of tactile information. Tactile wheels using analog tactile information perform better in terms of adaptability to unknown terrain.KeywordsIrregular terrainTactile wheelReinforcement learning
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
