A Novel Model for Robots to Avoid Obstacles based on Tensor Analysis and Differential Geometry
文献类型：会议论文
| 作者 | Shaokun Jin; Yongsheng Ou; Xinyu Wu; Wei Feng |
| 出版日期 | 2016 |
| 会议名称 | IEEE International Conference on Real-time Computing and Robotics (RCAR) |
| 会议地点 | 柬埔寨吴哥窟 |
| 英文摘要 | This paper attempts to build a novel model for robots to undertake obstacle avoidance tasks. Conventional work mainly forms a dynamical system (DS) through learning. The dynamical system learned from given demonstrations thus control the robot to perform tasks as expected. However, when the task becomes more complex, the form of dynamical system turns into higher nonlinear form. This results in huge time consumption and inaccuracy. In this paper, we systematically construct the mathematical model of the motion of the robot arm (including both links and the end-effector) based on which we analyze the policy for the robot to avoid stochastically interspersed obstacles and to approach targets. Tensor analysis is utilized to compute the gradient tensor of the distance between obstacles and links for robot to effectively avoid the obstacles. The Frenet frame is applied to model the trajectory of the end-effector and some other knowledge of differential geometry is involved for planning the trajectory of the endeffector to approach targets with avoiding all the obstacles. Several instances are provided to display the effectiveness of the novel model. |
| 收录类别 | EI |
| 语种 | 英语 |
| 源URL | [http://ir.siat.ac.cn:8080/handle/172644/10146]  |
| 专题 | 深圳先进技术研究院_集成所 |
| 作者单位 | 2016 |
| 推荐引用方式 GB/T 7714 | Shaokun Jin,Yongsheng Ou,Xinyu Wu,et al. A Novel Model for Robots to Avoid Obstacles based on Tensor Analysis and Differential Geometry[C]. 见:IEEE International Conference on Real-time Computing and Robotics (RCAR). 柬埔寨吴哥窟. |
入库方式： OAI收割
来源：深圳先进技术研究院
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。