4 papers
ContractionPPO: Certified Reinforcement Learning via Differentiable Contraction Layers
Vrushabh Zinage, Narek Harutyunyan, Eric Verheyden +2
Legged locomotion in unstructured environments demands not only high-performance control policies but also formal guarantees to ensure robustness under perturbations. Control metho…
Universal Barrier Functions for Safety and Stability of Constrained Nonlinear Systems
Vrushabh Zinage, Efstathios Bakolas
In this paper, we address the problem of synthesizing safe and stabilizing controllers for nonlinear systems subject to complex safety specifications and input constraints. We intr…
Deadlock-free, Safe, and Decentralized Multi-Robot Navigation in Social Mini-Games via Discrete-Time Control Barrier Functions
Rohan Chandra, Vrushabh Zinage, Efstathios Bakolas +2
We present an approach to ensure safe and deadlock-free navigation for decentralized multi-robot systems operating in constrained environments, including doorways and intersections…
Transformer-Based Model Predictive Path Integral Control
Shrenik Zinage, Vrushabh Zinage, Efstathios Bakolas
This paper presents a novel approach to improve the Model Predictive Path Integral (MPPI) control by using a transformer to initialize the mean control sequence. Traditional MPPI m…