We propose a Stein variational distributionally robust controller for nonlinear dynamical systems with latent parametric uncertainty. The method is an alternative to conservative worst-case ambiguity-set optimization with a deterministic particle-based approximation of a task-dependent uncertainty distribution, enabling the controller to concentrate on parameter sensitivities that most strongly affect closed-loop performance. Our method yields a controller that is robust to latent parameter uncertainty by coupling optimal control with Stein variational inference, and avoiding restrictive parametric assumptions on the uncertainty model while preserving computational parallelism. In contrast to classical DRO, which can sacrifice nominal performance through worst-case design, we find our approach achieves robustness by shaping the control law around relevant uncertainty that are most critical to the task objective. The proposed framework therefore reconciles robust control and variational inference in a single decision-theoretic formulation for broad classes of control systems with parameter uncertainty. We demonstrate our approach on representative control problems that empirically illustrate improved performance-robustness tradeoffs over nominal, ensemble, and classical distributionally robust baselines.
Georgia Tech（佐治亚理工学院）；NVIDIA（英伟达）；University of Washington（华盛顿大学）；The University of Sydney（悉尼大学）
Decision making under uncertainty is critical to real-world, autonomous systems. Model Predictive Control (MPC) methods have demonstrated favorable performance in practice, but remain limited when dealing with complex probability distributions. In this paper, we propose a generalization of MPC that represents a multitude of solutions as posterior distributions. By casting MPC as a Bayesian inference problem, we employ variational methods for posterior computation, naturally encoding the complexity and multi-modality of the decision making problem. We present a Stein variational gradient descen…
Georgia Institute of Technology（佐治亚理工学院）；University of Washington（华盛顿大学）
Many Imitation and Reinforcement Learning approaches rely on the availability of expert-generated demonstrations for learning policies or value functions from data. Obtaining a reliable distribution of trajectories from motion planners is non-trivial, since it must broadly cover the space of states likely to be encountered during execution while also satisfying task-based constraints. We propose a sampling strategy based on variational inference to generate distributions of feasible, low-cost trajectories for high-dof motion planning tasks. This includes a distributed, particle-based motion pl…
STL-SVPIO: Signal Temporal Logic guided Stein Variational Path Integral Optimization
Department of Electrical and Systems Engineering, University of Pennsylvania（宾夕法尼亚大学电气与系统工程系）；Mechanical Engineering and Mechanics Department, Lehigh University（莱特大学机械工程与力学系）
STAMP: Differentiable Task and Motion Planning via Stein Variational Gradient Descent
University of Toronto（多伦多大学）；Carnegie Mellon University（卡内基梅隆大学）；NVIDIA（英伟达）；Massachusetts Institute of Technology（麻省理工学院）；Technical University of Darmstadt（达姆施塔特工业大学）；McGill University（麦吉尔大学）；University of Sydney（悉尼大学）
Stein-based Optimization of Sampling Distributions in Model Predictive Path Integral Control
Laboratory for Perceptive RObotics and Grounded REasoning SystemS（感知机器人与基础推理系统实验室）；Robotics Department（机器人学系）；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）
University of Washington（华盛顿大学）；DEVCOM Army Research Laboratory（DEVCOM陆军研究实验室）
Efficient and reliable generation of global path plans are necessary for safe execution and deployment of autonomous systems. In order to generate planning graphs which adequately resolve the topology of a given environment, many sampling-based motion planners resort to coarse, heuristically-driven strategies which often fail to generalize to new and varied surroundings. Further, many of these approaches are not designed to contend with partial-observability. We posit that such uncertainty in environment geometry can, in fact, help drive the sampling process in generating feasible, and probabi…
SVN-ICP: Uncertainty Estimation of ICP-based LiDAR Odometry using Stein Variational Newton
Learning & Intelligent Systems Research Lab, TU Berlin（学习与智能系统研究实验室，柏林技术大学）；Technical University of Munich（慕尼黑技术大学）；Munich Institute of Robotics and Machine Intelligence (MIRMI)（慕尼黑机器人与机器智能研究所（MIRMI））
本文提出 SVN-ICP 算法，利用 Stein Variational Newton 在流形上实现基于 ICP 的激光雷达里程计不确定性估计，无需显式噪声建模，在多传感器融合中提供准确位姿估计和可靠不确定性。
Distributionally Robust Control via Stein Variational Inference for Contact-Rich Manipulation
Department of Mechanical Engineering, Yale University（耶鲁大学机械工程系）；Department of Electromechanical, Systems and Metal Engineering, Ghent University（根特大学机电系统与金属工程系）；School of Interactive Computing, Georgia Institute of Technology（佐治亚理工学院交互计算学院）；Department of Electrical Engineering, University of Sydney（悉尼大学电子工程系）
PinPoint: Monocular Needle Pose Estimation for Robotic Suturing via Stein Variational Newton and Geometric Residuals
Department of Mechanical Engineering, Vanderbilt University（范德比尔特大学机械工程系）；Department of Electrical and Computer Engineering and Department of Computer Science, Vanderbilt University（范德比尔特大学电气与计算机工程系和计算机科学系）；The Robotics Center and the Kahlert School of Computing, University of Utah（犹他大学机器人中心和Kahlert计算学院）
VIMPPI: Enhancing Model Predictive Path Integral Control with Variational Integration for Underactuated Systems
Interactive Robotic Systems Laboratory, Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院交互机器人系统实验室）；Institute of Robotics and Computer Vision, Innopolis University（伊诺波利斯大学机器人与计算机视觉研究所）
Korea Advanced Institute of Science and Technology（韩国科学技术院）；Okinawa Institute of Science and Technology（冲绳科学技术大学院大学）
Mental simulation is a critical cognitive function for goal-directed behavior because it is essential for assessing actions and their consequences. When a self-generated or externally specified goal is given, a sequence of actions that is most likely to attain that goal is selected among other candidates via mental simulation. Therefore, better mental simulation leads to better goal-directed action planning. However, developing a mental simulation model is challenging because it requires knowledge of self and the environment. The current paper studies how adequate goal-directed action plans of…
