News Release

IROS 2021 preview: robotic mapping and manipulation

Sept. 22, 2021-- Researchers at the UC San Diego Contextual Robotics Institute will present three papers at the IROS 2021 conference, which takes place Sept. 27 to Oct. 1, 2021, both online and in person in Prague. 

The papers focus on mapping and exploration for mobile robots and dexterous robotic manipulation. 

Active Exploration and Mapping via Iterative Covariance Regulation over Continuous SE(3) Trajectories

Shumon Koga, Arash Asgharivaskasi, Nikolay Atanasov

This paper develops \emph{iterative Covariance Regulation} (iCR), a novel method for active exploration and mapping for a mobile robot equipped with on-board sensors. The problem is posed as optimal control over the SE(3) pose kinematics of the robot to minimize the differential entropy of the map conditioned the potential sensor observations. Researchers introduce a differentiable field of view formulation, and derive iCR via the gradient descent method to iteratively update an open-loop control sequence in continuous space so that the covariance of the map estimate is minimized. The team demonstrates autonomous exploration and uncertainty reduction in simulated occupancy grid environments.

CORSAIR: Convolutional Object Retrieval and Symmetry-AIded Registration

Tianyu Zhao, Qiaojun Feng, Sai Jadhav, Nikolay Atanasov

This paper considers online object-level mapping using partial point-cloud observations obtained online in an unknown environment. Researchers develop an approach for fully Convolutional Object Retrieval and Symmetry-AIded Registration (CORSAIR). The model extends the Fully Convolutional Geometric Features model to learn a global object-shape embedding in addition to local point-wise features from the point-cloud observations. The global feature is used to retrieve a similar object from a category database, and the local features are used for robust pose registration between the observed and the retrieved object. The formulation also leverages symmetries, present in the object shapes, to obtain promising local-feature pairs from different symmetry classes for matching. Researchers present results from synthetic and real-world datasets with different object categories to verify the robustness of our method.

State-Only Imitation Learning for Dexterous Manipulation

Ilija Radosavovic1, Xiaolong Wang2, Lerrel Pinto3, Jitendra Malik1

Modern model-free reinforcement learning methods have recently demonstrated impressive results on a number of problems. However, complex domains like dexterous manipulation remain a challenge due to the high sample complexity. To address this, current approaches employ expert demonstrations in the form of state-action pairs, which are difficult to obtain for real-world settings such as learning from videos. In this paper, researchers move toward a more realistic setting and explore state-only imitation learning. To tackle this setting, the team trained an inverse dynamics model and used it to predict actions for state-only demonstrations. The inverse dynamics model and the policy are trained jointly. The method performs on par with state-action approaches and considerably outperforms RL alone. By not relying on expert actions, researchers are able to learn from demonstrations with different dynamics, morphologies, and objects.


Media Contacts

Ioana Patringenaru
Jacobs School of Engineering