.png){kind=icon}
返回
Safe, Optimal, Real-Time Trajectory Planning With a Parallel Constrained Bernstein Algorithm
.png){kind=icon}
OA
AI
DOI：10.1109/TRO.2020.3036617.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
To move while using new sensor information, mobile robots use receding-horizon planning, executing a short plan while computing a new one. A plan should have dynamic feasibility (obeying a robot's dynamics and avoiding obstacles), liveness (planning frequently enough to complete tasks), and optimality (minimizing, e.g., distance to a goal). Reachability-based trajectory design (RTD) is a method to generate provably dynamically feasible plans in real time by solving a polynomial optimization program (POP) in each planning iteration. However, RTD uses a derivative-based solver, which may converge to local minima that impact liveness and optimality. This article proposes a parallel constrained Bernstein algorithm (PCBA) branch-and-bound method to optimally solve RTD's POP at runtime; the resulting optimal planner is called RTD*. The specific contributions of this article are the PCBA implementation, proofs of PCBA's bounded time and memory usage, a comparison of PCBA with state-of-the-art solvers, and a demonstration of PCBA/RTD* on hardware. RTD* shows better optimality and liveness than RTD in dozens of environments with random obstacles.
Keyword:
Robots
Planning
Robot sensing systems
Real-time systems
Cost function
Trajectory planning
Heuristic algorithms
Motion planning
robot control
trajectory optimization
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
10.5
论文数:
3.3K
被引数:
2.8W
机构
引用论文
.png){kind=icon}
分享收藏Advancement of Nanobiomaterials to Deliver Natural Compounds for Tissue Engineering ApplicationsOAAI
OAAI分享收藏分享收藏
查看更多内容