|
轨迹预测：学习将情况映射到机器人轨迹
Trajectory Prediction: Learning to Map Situations to Robot Trajectories |
|
| |
| 课程网址: |
https://videolectures.net/videos/icml09_jetchev_tpl
|
| 主讲教师: |
Nikolay Jetchev
|
| 开课单位: |
会议
|
| 开课时间: |
2009-08-26
|
| 课程语种: |
英语
|
| 中文简介: |
轨迹规划和优化是关节机器人的一个基本问题。通常用于此问题的算法在新情况下从头开始计算最优轨迹。实际上，积累了大量数据，其中包含各种情况以及相应的优化轨迹，但这些数据在实践中几乎没有被利用。本文的目的是从这些数据中学习。考虑到一种新的情况，我们想预测一个合适的轨迹，只需要传统优化器进行微调。我们的方法有两个基本要素。首先，为了从以前的情况推广到新的情况，我们需要一个适当的情况描述符——我们提出了一种稀疏特征选择方法来找到这种情况的良好推广特征。其次，不应在关节角度空间中将先前优化的轨迹转移到新情况——我们提出了一种将旧轨迹更有效地转移到新状态的任务空间。模拟类人到达问题的轨迹优化实验表明，我们可以在新的情况下预测合理的运动原型，对于这些情况，改进比从头开始的优化快得多。
|
| 课程简介: |
Trajectory planning and optimization is a fundamental problem in articulated robotics. Algorithms used typically for this problem compute optimal trajectories from scratch in a new situation. In effect, extensive data is accumulated containing situations together with the respective optimized trajectories - but this data is in practice hardly exploited. The aim of this paper is to learn from this data. Given a new situation we want to predict a suitable trajectory which only needs minor refinement by a conventional optimizer. Our approach has two essential ingredients. First, to generalize from previous situations to new ones we need an appropriate situation descriptor - we propose a sparse feature selection approach to find such well-generalizing features of situations. Second, the transfer of previously optimized trajectories to a new situation should not be made in joint angle space - we propose a more efficient task space transfer of old trajectories to new situations. Experiments on trajectory optimization for a simulated humanoid reaching problem show that we can predict reasonable motion prototypes in new situations for which the refinement is much faster than an optimization from scratch.
|
| 关 键 词: |
轨迹预测; 机器人轨迹; 轨迹规划
|
| 课程来源: |
视频讲座网
|
| 数据采集: |
2025-04-25：liyq
|
| 最后编审: |
2025-04-25：liyq
|
| 阅读次数: |
5
|
|
图 书 馆