# Shared Autonomy via Hindsight Optimization for Teleoperation and Teaming

## TL;DR

This paper introduces a POMDP-based framework for shared autonomy that provides more effective assistance during teleoperation and teaming by minimizing expected cost-to-go, outperforming traditional predict-then-act methods.

## Contribution

The paper formalizes shared autonomy as a POMDP and applies hindsight optimization to improve assistance in teleoperation and teaming scenarios.

## Key findings

- Faster goal achievement compared to predict-then-act methods
- Requires less user input and reduces user idling time
- Fewer user-robot collisions observed

## Abstract

In shared autonomy, a user and autonomous system work together to achieve shared goals. To collaborate effectively, the autonomous system must know the user's goal. As such, most prior works follow a predict-then-act model, first predicting the user's goal with high confidence, then assisting given that goal. Unfortunately, confidently predicting the user's goal may not be possible until they have nearly achieved it, causing predict-then-act methods to provide little assistance. However, the system can often provide useful assistance even when confidence for any single goal is low (e.g. move towards multiple goals). In this work, we formalize this insight by modelling shared autonomy as a Partially Observable Markov Decision Process (POMDP), providing assistance that minimizes the expected cost-to-go with an unknown goal. As solving this POMDP optimally is intractable, we use hindsight optimization to approximate. We apply our framework to both shared-control teleoperation and human-robot teaming. Compared to predict-then-act methods, our method achieves goals faster, requires less user input, decreases user idling time, and results in fewer user-robot collisions.