Enabling Robots to Communicate Their Objectives

2017/02/28 by Sandy H. Huang, David Held, Pieter Abbeel +1 · 30 citations
Computer Science · Engineering · #Artificial intelligence #Behavior-based robotics #Computer science #Function (biology) #Human–computer interaction #Inference #Machine Learning and Algorithms #Machine learning #Mobile robot #Process (computing) #Reinforcement Learning in Robotics #Robot #Robot Manipulation and Learning #Robot learning #cs.LG #cs.RO

paper · pdf · doi:10.15607/rss.2017.xiii.059

RSS 2017

openalex publication_date 2017/07/12 · arxiv created 2018/10/18 · arxiv updated 2018/10/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

The overarching goal of this work is to efficiently enable end-users to correctly anticipate a robot's behavior in novel situations. Since a robot's behavior is often a direct result of its underlying objective function, our insight is that end-users need to have an accurate mental model of this objective function in order to understand and predict what the robot will do.

Citations

• Near-Optimally Teaching the Crowd to Classify
• An analysis of approximations for maximizing submodular set functions—I
• Continuous Inverse Optimal Control with Locally Optimal Examples
• Artificial intelligence: a modern approach

Cited by

• Surrogate Fitness Metrics for Interpretable Reinforcement Learning
• Towards Human Modeling for Human-Robot Collaboration and Digital Twins in Industrial Environments: Research Status, Prospects, and Challenges
• Counterfactual state explanations for reinforcement learning agents via generative deep learning
• Interactive Explanations for Reinforcement-Learning Agents