A survey of optimization-based task and motion planning: From classical to learning approaches
Task and motion planning (TAMP) integrates high-level task planning and low-level motion
planning to equip robots with the autonomy to effectively reason over long-horizon, dynamic …
planning to equip robots with the autonomy to effectively reason over long-horizon, dynamic …
Optimization-based control for dynamic legged robots
In a world designed for legs, quadrupeds, bipeds, and humanoids have the opportunity to
impact emerging robotics applications from logistics, to agriculture, to home assistance. The …
impact emerging robotics applications from logistics, to agriculture, to home assistance. The …
Robust multi-agent reinforcement learning via adversarial regularization: Theoretical foundation and stable algorithms
Abstract Multi-Agent Reinforcement Learning (MARL) has shown promising results across
several domains. Despite this promise, MARL policies often lack robustness and are …
several domains. Despite this promise, MARL policies often lack robustness and are …
Benefits of overparameterized convolutional residual networks: Function approximation under smoothness constraint
Overparameterized neural networks enjoy great representation power on complex data, and
more importantly yield sufficiently smooth output, which is crucial to their generalization and …
more importantly yield sufficiently smooth output, which is crucial to their generalization and …
Learning generalizable vision-tactile robotic grasping strategy for deformable objects via transformer
Reliable robotic grasping, especially with deformable objects such as fruits, remains a
challenging task due to underactuated contact interactions with a gripper, unknown object …
challenging task due to underactuated contact interactions with a gripper, unknown object …
Real-time deformable-contact-aware model predictive control for force-modulated manipulation
The force modulation of robotic manipulators has been extensively studied for several
decades. However, it is not yet commonly used in safety-critical applications due to a lack of …
decades. However, it is not yet commonly used in safety-critical applications due to a lack of …
Initial value problem enhanced sampling for closed-loop optimal control design with deep neural networks
Closed-loop optimal control design for high-dimensional nonlinear systems has been a long-
standing challenge. Traditional methods, such as solving the associated Hamilton-Jacobi …
standing challenge. Traditional methods, such as solving the associated Hamilton-Jacobi …
Trade-Off Between Robustness and Rewards Adversarial Training for Deep Reinforcement Learning Under Large Perturbations
J Huang, HJ Choi, N Figueroa - IEEE Robotics and Automation …, 2023 - ieeexplore.ieee.org
Deep Reinforcement Learning (DRL) has become a popular approach for training robots
due to its generalization promise, complex task capacity and minimal human intervention …
due to its generalization promise, complex task capacity and minimal human intervention …
Distributional Cloning for Stabilized Imitation Learning via ADMM
The two leading solution paradigms for imitation learning (IL), BC and GAIL, each suffers
from notable drawbacks. BC, a supervised learning approach to mimic expert actions, is …
from notable drawbacks. BC, a supervised learning approach to mimic expert actions, is …
LipsNet: a smooth and robust neural network with adaptive Lipschitz constant for high accuracy optimal control
Deep reinforcement learning (RL) is a powerful approach for solving optimal control
problems. However, RL-trained policies often suffer from the action fluctuation problem …
problems. However, RL-trained policies often suffer from the action fluctuation problem …