A neuro-symbolic framework for friction learning and physics-infused off-road navigation.

A neuro-symbolic framework for friction learning and physics-infused off-road navigation.

A self-supervised path-planning method to imporve the search efficiency of A* algorithm.

A vision-to-control humanoid stepping controller enhanced by Imperative Learning

A neural-symbolic motion prediction model integrating the conservation law into neural networks

An efficient point-line vSLAM addressing both short-term and long-term illumination challengs.

A self-supervised approach to learn feature matching

Imperative Learning is a Self-supervised Neural-Symbolic Learning Framework for Robot Autonomy

A Self-supervised Approach to Efficiently Solve Min-Max MTSP

An extremly efficient and accurate SLAM solution for warehouse robots.

The first 3D-geometry aware instance detector identifing novel instances in the wild.

The first self-supervised SLAM framework using imperative learning.

The first self-supervised path planning system using imperative learning.

Depth completion is crucial for many robotic tasks such a...

A simple yet effective object-based indoor relocalization approach.

An efficient learnable edge-based line detection method with local hough voting.

Autonomous exploration has many important applications. H...

An Illumination-Robust Point-Line Visual Odometry

One of the first temporal object encoding methods.

The first graph model converting a dynamic graph learning problem to lifelong learning.

A new pipeline, named AirDOS, for stereo Dynamic Object SLAM

Air Series is a collection of articles that are first authored by junior researchers.

A simple yet effective pipeline for robust object encoding with sparse encoding.

One of the first work to study loop closure detection in the lifelong learning context.

One of the first semantic RGB-D SLAM that run in real-time on a low-power embedded platform.

One of the most accurate and fastest open-sourced SLAM systems in KITTI dataset ranking.

Visual Memorability for Robotic Interestingness via Unsupervised Online Learning

Best paper award in robotic planning in the International Conference on Advanced Robotics (ICAR).