6 papers
Beyond Imitation: Self-Improving Robot Policies via Off-Policy Q-Planning
Varun Giridhar, Anant Khandelwal, Jeremy A. Collins +2
Behaviour Cloning (BC) has driven remarkable progress in robot manipulation, yet it is fundamentally limited by its inability to self-improve: a policy that fails cannot learn from…
DiffSim2Real: Deploying Quadrupedal Locomotion Policies Purely Trained in Differentiable Simulation
Joshua Bagajo, Clemens Schwarke, Victor Klemm +5
Differentiable simulators provide analytic gradients, enabling more sample-efficient learning algorithms and paving the way for data intensive learning tasks such as learning from…
PWM: Policy Learning with Multi-Task World Models
Ignat Georgiev, Varun Giridhar, Nicklas Hansen +1
Reinforcement Learning (RL) has made significant strides in complex tasks but struggles in multi-task settings with different embodiments. World model methods offer scalability by…
Adaptive Horizon Actor-Critic for Policy Learning in Contact-Rich Differentiable Simulation
Ignat Georgiev, Krishnan Srinivasan, Jie Xu +2
Model-Free Reinforcement Learning (MFRL), leveraging the policy gradient theorem, has demonstrated considerable success in continuous control tasks. However, these approaches are p…
Learning Deployable Locomotion Control via Differentiable Simulation
Clemens Schwarke, Victor Klemm, Joshua Bagajo +4
Differentiable simulators promise to improve sample efficiency in robot learning by providing analytic gradients of the system dynamics. Yet, their application to contact-rich task…
Iterative Semi-parametric Dynamics Model Learning For Autonomous Racing
Ignat Georgiev, Christoforos Chatzikomis, Timo Völkl +2
Accurately modeling robot dynamics is crucial to safe and efficient motion control. In this paper, we develop and apply an iterative learning semi-parametric model, with a neural n…