Dynamical system learning using extreme learning machines with safety and stability guarantees
Dynamical system learning using extreme learning machines with safety and stability guarantees
复制标题
使用极限学习机进行动态系统学习，安全稳定有保证
DOI：
10.1002/acs.3237
复制
发表时间：
2021
影响因子：
3.1
通讯作者：
Ashwin P. Dani
中科院分区：
文献类型：
--
作者：
Iman Salehi;G. Rotithor;Gang Yao;Ashwin P. Dani
本文提出了一种连续动态系统模型学习方法，可用于生成自治系统遵循的参考轨迹，使得这些轨迹对于给定的闭集是不变的，并且最终一致地关于闭集内的平衡点有界。自主系统动力学使用极限学习机 (ELM) 进行近似，其参数是根据使用倒数障碍函数表示的安全约束以及在存在 ELM 重建误差的情况下使用 Lyapunov 分析导出的稳定性约束来学习的。该公式可求解约束二次规划 (QP)，其中包含有限数量的决策变量和无限数量的约束。开发了一些定理，将具有无限数量约束的 QP 放松为具有有限数量约束的 QP，这实际上可以使用 QP 求解器来实现。此外，还开发了一种主动采样方法，通过仅评估较小点子集的约束，进一步减少了 QP 所需的约束数量。使用七自由度巴克斯特机器人上的运动再现任务来验证所提出的方法，其中使用所提出的方法来学习任务空间位置和速度动力学。
This article presents a continuous dynamical system model learning methodology that can be used to generate reference trajectories for the autonomous systems to follow, such that these trajectories are invariant to a given closed set and uniformly ultimately bounded with respect to an equilibrium point inside the closed set. The autonomous system dynamics are approximated using extreme learning machines (ELM), the parameters of which are learned subject to the safety constraints expressed using a reciprocal barrier function, and the stability constraints derived using a Lyapunov analysis in the presence of the ELM reconstruction error. This formulation leads to solving a constrained quadratic program (QP) that includes a finite number of decision variables with an infinite number of constraints. Theorems are developed to relax the QP with infinite number of constraints to a QP with a finite number of constraints which can be practically implemented using a QP solver. In addition, an active sampling methodology is developed that further reduced the number of required constraints for the QP by only evaluating the constraints at a smaller subset of points. The proposed method is validated using a motion reproduction task on a seven degree‐of‐freedom Baxter robot, where the task space position and velocity dynamics are learned using the presented methodology.
DOI：
10.23919/acc.2019.8815335
发表时间：
2019-07
期刊：
2019 American Control Conference (ACC)
影响因子：
--
作者：
Yongliang Yang;Yixin Yin;Wei He;K. Vamvoudakis;H. Modares;D. Wunsch
通讯作者：
Yongliang Yang;Yixin Yin;Wei He;K. Vamvoudakis;H. Modares;D. Wunsch
DOI：
10.1109/iros45743.2020.9341190
发表时间：
2020-03
期刊：
2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
影响因子：
--
作者：
Mohit Srinivasan;A. Dabholkar;S. Coogan;P. Vela
通讯作者：
Mohit Srinivasan;A. Dabholkar;S. Coogan;P. Vela