Shared Autonomy via Hindsight Optimization
This addresses the challenge of efficient human-robot collaboration in shared autonomy, but it is incremental as it builds on existing predict-then-blend approaches with a new approximation method.
The paper tackled the problem of shared autonomy where a robot must predict and assist with a user's unknown goal, using a POMDP formulation and hindsight optimization to approximate optimal actions. In a user study, the method enabled users to accomplish tasks more quickly with less input, though user assessments were mixed due to a tradeoff between control authority and speed.
In shared autonomy, user input and robot autonomy are combined to control a robot to achieve a goal. Often, the robot does not know a priori which goal the user wants to achieve, and must both predict the user's intended goal, and assist in achieving that goal. We formulate the problem of shared autonomy as a Partially Observable Markov Decision Process with uncertainty over the user's goal. We utilize maximum entropy inverse optimal control to estimate a distribution over the user's goal based on the history of inputs. Ideally, the robot assists the user by solving for an action which minimizes the expected cost-to-go for the (unknown) goal. As solving the POMDP to select the optimal action is intractable, we use hindsight optimization to approximate the solution. In a user study, we compare our method to a standard predict-then-blend approach. We find that our method enables users to accomplish tasks more quickly while utilizing less input. However, when asked to rate each system, users were mixed in their assessment, citing a tradeoff between maintaining control authority and accomplishing tasks quickly.