文本课件教程slam关键.pdf
- wx5620个人认证 |
- 2021-12-09 发布|
- 2.61 MB|
- 7页
2017 IEEE International Conference on Robotics and Automation (ICRA)
Singapore, May 29 - June 3, 2017 Leveraging Experience for Computationally Efficient Adaptive Nonlinear M Predictive Control Vishnu R. Desaraju and Nathan Michael Abstract— This work presents Experience-driven Predictive reactive techniques that seek to eliminate the effects of
Control (EPC) as a fast technique for solving nonlinear m unm ed dynamics, even when they may be beneficial. In
predictive control (NMPC) problems with uncertain system contrast, m predictive control (MPC) techniques seek to
dynamics. EPC leverages an affine dynamics m that is balance the reactive nature of traditional feedback controllers
updated online via Locally Weighted Projection Regression
(LWPR) to capture nonlinearities, uncertainty, and chan