ICRA 2020

494 papers
The 2020 International Conference on Robotics and Automation featured research focused on medical applications, safety in control systems, and multi-agent coordination. Contributions in the medical sector included studies on post-stroke rehabilitation through temporal muscle synergy analysis and the development of semi-autonomous robots for stereotactic brain biopsies. These works addressed the integration of robotic assistance in clinical settings to improve patient outcomes and procedural safety. In the domain of control and optimization, research addressed safe optimal control under parametric uncertainty and the use of motion memory to facilitate warm-start trajectory optimization. Additionally, the conference highlighted advancements in multi-robot systems, specifically regarding the scalable cooperative transport of cable-suspended loads using unmanned aerial vehicles (UAVs). This research utilized distributed trajectory optimization to manage complex aerial maneuvers. Prolific contributors to these themes included researchers such as An Qi, Atsushi Yamashita, and Hajime Asama, reflecting an emphasis on refining autonomous navigation and human-robot interaction through algorithmic improvements.

Top Research Directions

Research directions most represented by ICRA 2020 papers.

Zero-Shot Sim-to-Real Transfer for Robotic Control 8
Resolving Geometric Symmetry Ambiguity in 3D Pose Estimation 5
Sample Efficient Reinforcement Learning 5
Overcoming the Reality Gap in Robotic Control 5
Generalizing from Synthetic Data to the Real World 4
Inferring Latent Physical Properties from Object Interactions 3
Interaction Aware Planning for Overcoming Navigation Paralysis 3
Integrating Social Norms into Robot Motion Planning 3
Scaling Robotic Control to Long-Horizon Multi-Stage Tasks 3
Mitigating Dynamic Object Interference in Visual SLAM 3
Motion Planning Beyond the Static Environment Assumption 3
Overcoming Depth Sensor Failures on Transparent Surfaces 3
Safe Reinforcement Learning via Constrained Markov Decision Processes 3
Accelerating Trajectory Optimization for Real-Time Robotic Control 3
Scaling Robotic Control to High Dimensional Action Spaces 3

Top Authors

AuthorPapers
Dieter Fox12
Alois Knoll5
Dinesh Manocha5
Fabio Ramos5
H. Jin Kim5
Shaojie Shen5
Animesh Garg4
Claire Tomlin4
Daniela Rus4
Davide Scaramuzza4
Jeannette Bohg4
Jeffrey Ichnowski4
Jia Pan4
Ken Goldberg4
Li Fei-Fei4
Marco Pavone4
Matthew Johnson-Roberson4
Nikolay Atanasov4
Andy Zeng3
Angela P. Schoellig3

Impactful Papers

Papers from ICRA 2020 that shaped research clusters.

Foundational

Influential

Papers

Title
Learned Sampling Distributions for Efficient Planning in Hybrid Geometric and Object-Level Representations
Learning 3D-aware Egocentric Spatial-Temporal Interaction via Graph Convolutional Networks
Learning a Control Policy for Fall Prevention on an Assistive Walking Device
Learning Affordance Space in Physical World for Vision-based Robotic Object Manipulation
Learning an Action-Conditional Model for Haptic Texture Generation
Learning a Pile Loading Controller from Demonstrations
Learning Camera Miscalibration Detection
Learning Control Policies from Optimal Trajectories
Learning Face Recognition Unsupervisedly by Disentanglement and Self-Augmentation
Learning from Simulation, Racing in Reality
Learning Generalizable Locomotion Skills with Hierarchical Reinforcement Learning
Learning Grasping Points for Garment Manipulation in Robot-Assisted Dressing
Learning Hierarchical Control for Robust In-Hand Manipulation
Learning How to Walk: Warm-starting Optimal Control Solver with Memory of Motion
Learning local behavioral sequences to better infer non-local properties in real multi-robot systems
Learning Multi-Arm Manipulation Through Collaborative Teleoperation
Learning Multi-Robot Decentralized Macro-Action-Based Policies via a Centralized Q-Net
Learning Navigation Costs from Demonstration in Partially Observable Environments
Learning Precise 3D Manipulation from Multiple Uncalibrated Cameras
Learning Pregrasp Manipulation of Objects from Ungraspable Poses
Learning Resilient Behaviors for Navigation Under Uncertainty
Learning Robotic Assembly Tasks with Lower Dimensional Systems by Leveraging Physical Softness and Environmental Constraints
Learning Rope Manipulation Policies Using Dense Object Descriptors Trained on Synthetic Depth Data
Learning Shape-based Representation for Visual Localization in Extremely Changing Conditions
Learning Stable Normalizing-Flow Control for Robotic Manipulation
Learning Task Space Actions for Bipedal Locomotion
Learning to combine primitive skills: A step towards versatile robotic manipulation §
Learning to Drive Off Road on Smooth Terrain in Unstructured Environments Using an On-Board Camera and Sparse Aerial Images
Learning to Generate 6-DoF Grasp Poses with Reachability Awareness
Learning to Scaffold the Development of Robotic Manipulation Skills
Learning to See before Learning to Act: Visual Pre-training for Manipulation
Learning to Transfer Dynamic Models of Underactuated Soft Robotic Hands
Learning Whole-Body Human-Robot Haptic Interaction in Social Contexts
Least-squares Optimal Relative Planar Motion for Vehicle-mounted Cameras
Leveraging Neural Network Gradients within Trajectory Optimization for Proactive Human-Robot Interactions
Leveraging the Template and Anchor Framework for Safe, Online Robotic Gait Design
LiDAR-enhanced Structure-from-Motion
Lidar-Monocular Visual Odometry using Point and Line Features
Linear RGB-D SLAM for Atlanta World
LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery
Localising Faster: Efficient and precise lidar-based robot localisation in large-scale environments
Local Policy Optimization for Trajectory-Centric Reinforcement Learning
Local Trajectory Stabilization for Dexterous Manipulation via Piecewise Affine Approximations
Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Pooling
Long-Horizon Prediction and Uncertainty Propagation with Residual Point Contact Learners
Look, Listen, and Act: Towards Audio-Visual Embodied Navigation
Low Latency And Low-Level Sensor Fusion For Automotive Use-Cases
LyRN (Lyapunov Reaching Network): A Real-Time Closed Loop approach from Monocular Vision
MagNet: Discovering Multi-agent Interaction Dynamics using Neural Network
MAMS-A*: Multi-Agent Multi-Scale A

« prev Page 6/10 next »