Glen Chou
YOU?
Author Swipe
View article: Robustly Constrained Dynamic Games for Uncertain Nonlinear Dynamics
Robustly Constrained Dynamic Games for Uncertain Nonlinear Dynamics Open
We propose a novel framework for robust dynamic games with nonlinear dynamics corrupted by state-dependent additive noise, and nonlinear agent-specific and shared constraints. Leveraging system-level synthesis (SLS), each agent designs a n…
View article: Improving Out-of-Distribution Generalization of Learned Dynamics by Learning Pseudometrics and Constraint Manifolds
Improving Out-of-Distribution Generalization of Learned Dynamics by Learning Pseudometrics and Constraint Manifolds Open
We propose a method for improving the prediction accuracy of learned robot dynamics models on out-of-distribution (OOD) states. We achieve this by leveraging two key sources of structure often present in robot dynamics: 1) sparsity, i.e., …
View article: Fighting Uncertainty with Gradients: Offline Reinforcement Learning via Diffusion Score Matching
Fighting Uncertainty with Gradients: Offline Reinforcement Learning via Diffusion Score Matching Open
Gradient-based methods enable efficient search capabilities in high dimensions. However, in order to apply them effectively in offline optimization paradigms such as offline Reinforcement Learning (RL) or Imitation Learning (IL), we requir…
View article: Synthesizing Stable Reduced-Order Visuomotor Policies for Nonlinear Systems via Sums-of-Squares Optimization
Synthesizing Stable Reduced-Order Visuomotor Policies for Nonlinear Systems via Sums-of-Squares Optimization Open
We present a method for synthesizing dynamic, reduced-order output-feedback polynomial control policies for control-affine nonlinear systems which guarantees runtime stability to a goal state, when using visual observations and a learned p…
View article: Data-Efficient Learning of Natural Language to Linear Temporal Logic Translators for Robot Task Specification
Data-Efficient Learning of Natural Language to Linear Temporal Logic Translators for Robot Task Specification Open
To make robots accessible to a broad audience, it is critical to endow them with the ability to take universal modes of communication, like commands given in natural language, and extract a concrete desired task specification, defined usin…
View article: Statistical Safety and Robustness Guarantees for Feedback Motion Planning of Unknown Underactuated Stochastic Systems
Statistical Safety and Robustness Guarantees for Feedback Motion Planning of Unknown Underactuated Stochastic Systems Open
We present a method for providing statistical guarantees on runtime safety and goal reachability for integrated planning and control of a class of systems with unknown nonlinear stochastic underactuated dynamics. Specifically, given a dyna…
View article: Safe Output Feedback Motion Planning from Images via Learned Perception Modules and Contraction Theory
Safe Output Feedback Motion Planning from Images via Learned Perception Modules and Contraction Theory Open
We present a motion planning algorithm for a class of uncertain control-affine nonlinear systems which guarantees runtime safety and goal reachability when using high-dimensional sensor measurements (e.g., RGB-D images) and a learned perce…
View article: Safe End-to-end Learning-based Robot Autonomy via Integrated Perception, Planning, and Control
Safe End-to-end Learning-based Robot Autonomy via Integrated Perception, Planning, and Control Open
Trustworthy robots must be able to complete tasks reliably while obeying safety constraints. While traditional methods for constrained motion planning and optimal control can achieve this if the environment is accurately modeled and the ta…
View article: Gaussian Process Constraint Learning for Scalable Chance-Constrained Motion Planning from Demonstrations
Gaussian Process Constraint Learning for Scalable Chance-Constrained Motion Planning from Demonstrations Open
We propose a method for learning constraints represented as Gaussian processes (GPs) from locally-optimal demonstrations. Our approach uses the Karush-Kuhn-Tucker (KKT) optimality conditions to determine where on the demonstrations the con…
View article: Correction to “Planning With Learned Dynamics: Probabilistic Guarantees on Safety and Reachability Via Lipschitz Constants” [Jul21 5129-5136]
Correction to “Planning With Learned Dynamics: Probabilistic Guarantees on Safety and Reachability Via Lipschitz Constants” [Jul21 5129-5136] Open
We wish to make the following corrections and clarifications to our manuscript [1]. For a version of the manuscript that has these changes integrated into the text, please see [2]. •In [1], the method is claimed to provide safety guarantee…
View article: Compositional safety rules for inter-triggering hybrid automata
Compositional safety rules for inter-triggering hybrid automata Open
Extended version of the conference paper with a supplementary appendix.
View article: Model Error Propagation via Learned Contraction Metrics for Safe Feedback Motion Planning of Unknown Systems
Model Error Propagation via Learned Contraction Metrics for Safe Feedback Motion Planning of Unknown Systems Open
We present a method for contraction-based feedback motion planning of locally incrementally exponentially stabilizable systems with unknown dynamics that provides probabilistic safety and reachability guarantees. Given a dynamics dataset, …
View article: Planning With Learned Dynamics: Probabilistic Guarantees on Safety and Reachability via Lipschitz Constants
Planning With Learned Dynamics: Probabilistic Guarantees on Safety and Reachability via Lipschitz Constants Open
We present a method for feedback motion planning of systems with unknown dynamics which provides probabilistic guarantees on safety, reachability, and goal stability. To find a domain in which a learned control-affine approximation of the …
View article: Inferring Obstacles and Path Validity from Visibility-Constrained Demonstrations
Inferring Obstacles and Path Validity from Visibility-Constrained Demonstrations Open
View article: Uncertainty-Aware Constraint Learning for Adaptive Safe Motion Planning from Demonstrations
Uncertainty-Aware Constraint Learning for Adaptive Safe Motion Planning from Demonstrations Open
We present a method for learning to satisfy uncertain constraints from demonstrations. Our method uses robust optimization to obtain a belief over the potentially infinite set of possible constraints consistent with the demonstrations, and…
View article: Planning with Learned Dynamics: Guaranteed Safety and Reachability via Lipschitz Constants.
Planning with Learned Dynamics: Guaranteed Safety and Reachability via Lipschitz Constants. Open
We present an approach for feedback motion planning of systems with unknown
dynamics which provides guarantees on safety, reachability, and stability about
the goal. Given a learned control-affine approximation of the true dynamics, we
est…
View article: Planning with Learned Dynamics: Probabilistic Guarantees on Safety and\n Reachability via Lipschitz Constants
Planning with Learned Dynamics: Probabilistic Guarantees on Safety and\n Reachability via Lipschitz Constants Open
We present a method for feedback motion planning of systems with unknown\ndynamics which provides probabilistic guarantees on safety, reachability, and\ngoal stability. To find a domain in which a learned control-affine\napproximation of t…
View article: Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations
Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations Open
We present a method for learning to perform multistage tasks from demonstrations by learning the logical structure and atomic propositions of a consistent linear temporal logic (LTL) formula.The learner is given successful but potentially …
View article: Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations
Explaining Multi-stage Tasks by Learning Temporal Logic Formulas from Suboptimal Demonstrations Open
We present a method for learning multi-stage tasks from demonstrations by learning the logical structure and atomic propositions of a consistent linear temporal logic (LTL) formula. The learner is given successful but potentially suboptima…
View article: Learning Constraints from Locally-Optimal Demonstrations under Cost Function Uncertainty
Learning Constraints from Locally-Optimal Demonstrations under Cost Function Uncertainty Open
We present an algorithm for learning parametric constraints from locally-optimal demonstrations, where the cost function being optimized is uncertain to the learner. Our method uses the Karush-Kuhn-Tucker (KKT) optimality conditions of the…
View article: Learning Parametric Constraints in High Dimensions from Demonstrations
Learning Parametric Constraints in High Dimensions from Demonstrations Open
We present a scalable algorithm for learning parametric constraints in high dimensions from safe expert demonstrations. To reduce the ill-posedness of the constraint recovery problem, our method uses hit-and-run sampling to generate lower …
View article: Learning Constraints from Demonstrations
Learning Constraints from Demonstrations Open
We extend the learning from demonstration paradigm by providing a method for learning unknown constraints shared across tasks, using demonstrations of the tasks, their cost functions, and knowledge of the system dynamics and control constr…
View article: Using Control Synthesis to Generate Corner Cases: A Case Study on Autonomous Driving
Using Control Synthesis to Generate Corner Cases: A Case Study on Autonomous Driving Open
This paper employs correct-by-construction control synthesis, in particular controlled invariant set computations, for falsification. Our hypothesis is that if it is possible to compute a “large enough” controlled invariant set either for …
View article: Incremental Segmentation of ARX Models
Incremental Segmentation of ARX Models Open
We consider the problem of incrementally segmenting auto-regressive models with exogenous inputs (ARX models) when the data is received sequentially at run-time. In particular, we extend a recently proposed dynamic programming based polyno…
View article: A Hybrid Framework for Multi-Vehicle Collision Avoidance
A Hybrid Framework for Multi-Vehicle Collision Avoidance Open
With the recent surge of interest in UAVs for civilian services, the importance of developing tractable multi-agent analysis techniques that provide safety and performance guarantees have drastically increased. Hamilton-Jacobi (HJ) reachab…
View article: Using Neural Networks for Fast Reachable Set Computations
Using Neural Networks for Fast Reachable Set Computations Open
View article: Using Neural Networks to Compute Approximate and Guaranteed Feasible Hamilton-Jacobi-Bellman PDE Solutions
Using Neural Networks to Compute Approximate and Guaranteed Feasible Hamilton-Jacobi-Bellman PDE Solutions Open
To sidestep the curse of dimensionality when computing solutions to Hamilton-Jacobi-Bellman partial differential equations (HJB PDE), we propose an algorithm that leverages a neural network to approximate the value function. We show that o…