Active Reward Learning for Co-Robotic Vision Based Exploration in Bandwidth Limited Environments

03/10/2020

∙

We present a novel POMDP problem formulation for a robot that must autonomously decide where to go to collect new and scientifically relevant images given a limited ability to communicate with its human operator. From this formulation we derive constraints and design principles for the observation model, reward model, and communication strategy of such a robot, exploring techniques to deal with the very high-dimensional observation space and scarcity of relevant training data. We introduce a novel active reward learning strategy based on making queries to help the robot minimize path "regret" online, and evaluate it for suitability in autonomous visual exploration through simulations. We demonstrate that, in some bandwidth-limited environments, this novel regret-based criterion enables the robotic explorer to collect up to 17

READ FULL TEXT

Active Reward Learning for Co-Robotic Vision Based Exploration in Bandwidth Limited Environments

Streaming Scene Maps for Co-Robotic Exploration in Bandwidth Limited Environments

Active Exploration for Robotic Manipulation

Batch Exploration with Examples for Scalable Robotic Reinforcement Learning

Online reinforcement learning with sparse rewards through an active inference capsule

Robust Robotic Control from Pixels using Contrastive Recurrent State-Space Models

Disentangled Planning and Control in Vision Based Robotics via Reward Machines

Energy-Constrained Active Exploration Under Incremental-Resolution Symbolic Perception

Active Reward Learning for Co-Robotic Vision Based Exploration in Bandwidth Limited Environments

Related Research

Streaming Scene Maps for Co-Robotic Exploration in Bandwidth Limited Environments

Active Exploration for Robotic Manipulation

Batch Exploration with Examples for Scalable Robotic Reinforcement Learning

Online reinforcement learning with sparse rewards through an active inference capsule

Robust Robotic Control from Pixels using Contrastive Recurrent State-Space Models

Disentangled Planning and Control in Vision Based Robotics via Reward Machines

Energy-Constrained Active Exploration Under Incremental-Resolution Symbolic Perception