A Risk-Averse Preview-based $Q$-Learning Algorithm: Application to Highway Driving of Autonomous Vehicles

no code implementations • 6 Dec 2021 • Majid Mazouchi, Subramanya Nageshrao, Hamidreza Modares

A risk assessment unit module is then presented that leverages the preview information provided by sensors along with a stochastic reachability module to assign reward values to the MDP states and update them as scenarios develop.

Autonomous Vehicles Q-Learning

Finite-time Koopman Identifier: A Unified Batch-online Learning Framework for Joint Learning of Koopman Structure and Parameters

no code implementations • 12 May 2021 • Majid Mazouchi, Subramanya Nageshrao, Hamidreza Modares

In this paper, a unified batch-online learning approach is introduced to learn a linear representation of nonlinear system dynamics using the Koopman operator.

Bayesian Optimization

A Convex Programming Approach to Data-Driven Risk-Averse Reinforcement Learning

no code implementations • 26 Mar 2021 • Yuzhen Han, Majid Mazouchi, Subramanya Nageshrao, Hamidreza Modares

This paper presents a model-free reinforcement learning (RL) algorithm to solve the risk-averse optimal control (RAOC) problem for discrete-time nonlinear systems.

reinforcement-learning Reinforcement Learning (RL)

Assured Learning-enabled Autonomy: A Metacognitive Reinforcement Learning Framework

no code implementations • 23 Mar 2021 • Aquib Mustafa, Majid Mazouchi, Subramanya Nageshrao, Hamidreza Modares

To guarantee performance while assuring satisfaction of safety constraints across variety of circumstances, an assured autonomous control framework is presented in this paper by empowering RL algorithms with metacognitive learning capabilities.

Decision Making reinforcement-learning +1

Towards Interpretable-AI Policies Induction using Evolutionary Nonlinear Decision Trees for Discrete Action Systems

no code implementations • 20 Sep 2020 • Yashesh Dhebar, Kalyanmoy Deb, Subramanya Nageshrao, Ling Zhu, Dimitar Filev

In this paper, we use a recently proposed nonlinear decision-tree (NLDT) approach to find a hierarchical set of control rules in an attempt to maximize the open-loop performance for approximating and explaining the pre-trained black-box DRL (oracle) agent using the labelled state-action dataset.

Bilevel Optimization

An online evolving framework for advancing reinforcement-learning based automated vehicle control

no code implementations • 15 Jun 2020 • Teawon Han, Subramanya Nageshrao, Dimitar P. Filev, Umit Ozguner

With the latest stochastic model and given criteria, the action-reviser module checks validity of the controller's chosen action by predicting future states.

Decision Making reinforcement-learning +1

Autonomous Highway Driving using Deep Reinforcement Learning

no code implementations • 29 Mar 2019 • Subramanya Nageshrao, Eric Tseng, Dimitar Filev

This may lead to a scenario that was not postulated in the design phase.

reinforcement-learning Reinforcement Learning (RL)