• Publications
  • Influence
CHOMP: Covariant Hamiltonian optimization for motion planning
tl;dr
In this paper, we present CHOMP (covariant Hamiltonian optimization for motion planning), a method for trajectory optimization invariant to reparametrization. Expand
  • 370
  • 55
A policy-blending formalism for shared control
tl;dr
In shared control teleoperation, the robot assists the user in accomplishing the desired task by attempting to predict their intent and augment their input, making teleoperation easier and more seamless. Expand
  • 171
  • 29
Legibility and predictability of robot motion
tl;dr
A key requirement for seamless human-robot collaboration is for the robot to make its intentions clear to its human collaborator. Expand
  • 341
  • 23
Cooperative Inverse Reinforcement Learning
tl;dr
We propose a formal definition of the value alignment problem as cooperative inverse reinforcement learning (CIRL). Expand
  • 212
  • 15
Toward seamless human-robot handovers
tl;dr
A handover is a complex collaboration, where actors coordinate in time and space to transfer control of an object. Expand
  • 179
  • 15
Generating Legible Motion
tl;dr
Our algorithm optimizes a legibility metric inspired by the psychology of action interpretation in humans, resulting in motion trajectories that purposefully deviate from what an observer would expect in order to better convey intent. Expand
  • 128
  • 10
Active Preference-Based Learning of Reward Functions
tl;dr
Our goal is to efficiently learn reward functions encoding a human’s preferences for how a dynamical system should act. Expand
  • 99
  • 10
Effects of Robot Motion on Human-Robot Collaboration
tl;dr
Most motion in robotics is purelyfifnctional, planned to achieve the goal and avoid collisions. Expand
  • 133
  • 8
DART: Noise Injection for Robust Imitation Learning
tl;dr
We propose a new algorithm, DART (Disturbances for Augmenting Robot Trajectories), that collects demonstrations with injected noise, and optimizes the noise level to approximate the error of the robot's trained policy during data collection. Expand
  • 67
  • 8
Planning for Autonomous Cars that Leverage Effects on Human Actions
tl;dr
We model driving in an environment with a human driven car as a dynamical system, in which the robot’s actions have immediate consequences on the state of the car, but also on human actions. Expand
  • 203
  • 7