Abstract
ABSTRACT The configuration space is a fundamental concept that is widely used in algorithmic robotics. Many applications in robotics, computer-aided design, and related areas can be reduced to computational problems in terms of configuration spaces. In this paper, we survey some of our recent work on solving two important challenges related to configuration spaces: how to efficiently compute an approximate representation of high-dimensional configuration spaces; and how to efficiently perform geometric proximity and motion planning queries in high-dimensional configuration spaces. We present new configuration space construction algorithms based on machine learning and geometric approximation techniques. These algorithms perform collision queries on many configuration samples. The collision query results are used to compute an approximate representation for the configuration space, which quickly converges to the exact configuration space. We also present parallel GPU-based algorithms to accelerate the performance of optimization and search computations in configuration spaces. In particular, we design efficient GPU-based parallel k -nearest neighbor and parallel collision detection algorithms and use these algorithms to accelerate motion planning.
Highlights
Intelligent robots are becoming increasingly important in both industry and everyday life
Research Robotics—Article consists of its position and orientation, and the configuration space is SE(3) if both rotation and translation are allowed, and R3 if only translation is allowed; and the configuration of an articulated object is the vector of all its joint angles
For motion planning of high-DOF robots, most of the practical methods are based on randomized algorithms, including probabilistic roadmap (PRM) [26] and rapidly exploring random tree (RRT) [27]
Summary
Intelligent robots are becoming increasingly important in both industry and everyday life. The PR2 robot from Willow Garage has been shown to assist people with severe physical disabilities such as quadriplegia [4]; and humanoid robots such as the HRP-4 can perform human-like actions, and can communicate with people using speech [5] In addition to their applications in industry and everyday life, modern intelligent robots can be helpful in other areas, including autonomous vehicles [6], medical and surgical intervention [7], emergency and disaster rescue [8], and military tasks [9]. (2) Many robotics applications require real-time planning in order to work reliably and efficiently in human environments with moving obstacles, but performing optimization in the computed representation for the configuration space can be time consuming. We provide parallel GPU-based algorithms to accelerate the optimization computations in the configuration space, which can allow for real-time planning computation in many challenging environments (Section 5)
Sign-in/Register to view highlights & summary 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.
