ICRA 2018

366 papers
The 2018 International Conference on Robotics and Automation (ICRA) featured research spanning autonomous navigation, medical robotics, and multi-agent systems. Contributions included developments in LiDAR-based vehicle localization and lane-keeping systems utilizing spiking neural networks. Research in medical robotics addressed marker-free hand-eye calibration for eye surgery using optical coherence tomography. Additionally, the conference highlighted advancements in human-robot interaction through augmented reality instructions for shared control of hand-held systems. Multi-robot coordination was a recurring theme, with papers addressing task allocation for collaborative adaptive sampling and pilot surveys. Localization and mapping remained central, with work on vector-field SLAM using ultra-wideband ranging and magnetic field fusion. Practical applications were also represented, such as the development of unmanned aerial vehicle systems for bridge inspection. Prolific contributors to this year's proceedings included Gaurav S. Sukhatme, Stephanie Kemna, and Henrik I. Christensen.

Top Research Directions

Research directions most represented by ICRA 2018 papers.

Probabilistic Trajectory Forecasting for Autonomous Systems 4
Bridging the Simulation to Reality Transfer Gap 4
Robust Robotic Grasping in Unstructured Environments 3
Extrinsic Dexterity in Robotic Manipulation 3
Closed Loop Visual Servoing for Precise Robotic Control 3
Beyond Planar Constraints in Robotic Manipulation 3
Bridging the Embodiment Gap in Robot Learning 3
Modeling Reciprocal Dynamics in Multi-Agent Trajectory Prediction 3
Class Agnostic Localization for Open World Object Detection 3
Model Free Generalizable 6D Object Pose Estimation 3
Safe Reinforcement Learning via Constrained Markov Decision Processes 3
Overcoming Visual Self Occlusion in Robotic Manipulation 3
Scaling Robotic Learning Beyond Expert Demonstrations 3
Scaling Motion Planning to High Dimensional Spaces 3
Zero-Shot Sim-to-Real Transfer for Robotic Control 3

Top Authors

AuthorPapers
Sergey Levine8
Daniela Rus7
Ian Reid6
Paul Newman6
Sertac Karaman6
Gaurav S. Sukhatme5
Marco Pavone5
Nicholas Roy5
Pieter Abbeel5
Alois Knoll4
Danica Kragic4
Howie Choset4
Ken Goldberg4
Mac Schwager4
Shaojie Shen4
Sven Behnke4
Alex Bewley3
Andrew Markham3
Byron Boots3
Dinesh Manocha3

Impactful Papers

Papers from ICRA 2018 that shaped research clusters.

Foundational

Influential

Papers

Title
SceneCut: Joint Geometric and Object Segmentation for Indoor Scenes
SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Control
SeDAR - Semantic Detection and Ranging: Humans can Localise without LiDAR, can Robots?
Self-Modifying Morphology Experiments with DyRET: Dynamic Robot for Embodied Testing
Self-Supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigation
Semantic Labeling of Indoor Environments from 3D RGB Maps
Semantic Segmentation from Limited Training Data
Semi-Dense Visual-Inertial Odometry and Mapping for Quadrotors with SWAP Constraints
Sensor-Based Reactive Symbolic Planning in Partially Known Environments
Shaping in Practice: Training Wheels to Learn Fast Hopping Directly in Hardware
Sim-to-Real Transfer of Robotic Control with Dynamics Randomization
Situation Assessment for Planning Lane Changes: Combining Recurrent Models and Prediction
SLAMBench2: Multi-Objective Head-to-Head Benchmarking for Visual SLAM.
Social Attention: Modeling Attention in Human Crowds.
Socially Compliant Navigation Through Raw Depth Inputs with Generative Adversarial Imitation Learning
Sparse Gaussian Processes on Matrix Lie Groups: A Unified Framework for Optimizing Continuous-Time Trajectories
Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image
SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud
SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud
Stabilizing Traffic with Autonomous Vehicles
StaticFusion: Background Reconstruction for Dense RGB-D SLAM in Dynamic Environments
Stiffness Variability in Jamming of Compliant Granules and a Case Study Application in Climbing Vertical Shafts
SuperDepth: Self-Supervised, Super-Resolved Monocular Depth Estimation
Surface Edge Explorer (see): Planning Next Best Views Directly from 3D Observations
Synthetically Trained Neural Networks for Learning Human-Readable Plans from Real-World Demonstrations
Task-Specific Sensor Planning for Robotic Assembly Tasks
Temporal Spatial Inverse Semantics for Robots Communicating with Humans
Tensegrity Robot Locomotion Under Limited Sensory Inputs via Deep Reinforcement Learning
Text2Action: Generative Adversarial Synthesis from Language to Action
The Method of PID parameters tuning for AMB system at high-frequency
The UNAV, a Wind-Powered UAV for Ocean Monitoring: Performance, Control and Validation
Time-Contrastive Networks: Self-Supervised Learning from Video
Topological Multi-Robot Belief Space Planning in Unknown Environments
Touch-Based Grasp Primitives for Soft Hands: Applications to Human-to-Robot Handover Tasks and Beyond
Towards Optimally Decentralized Multi-Robot Collision Avoidance via Deep Reinforcement Learning
Towards Restoring Locomotion for Paraplegics: Realizing Dynamically Stable Walking on Exoskeletons
Towards Robotic Eye Surgery: Marker-free, Online Hand-eye Calibration using Optical Coherence Tomography Images, to be appear.
Towards Understanding Object-Directed Actions: A Generative Model for Grounding Syntactic Categories of Speech Through Visual Perception
TrackerBots: Software in the Loop Study of Quad-Copter Robots for Locating Radio-tags in a 3D Space
Track, Then Decide: Category-Agnostic Vision-Based Multi-Object Tracking
Training a Binary Weight Object Detector by Knowledge Transfer for Autonomous Driving
Trajectory-Optimized Sensing for Active Search of Tissue Abnormalities in Robotic Surgery
Trajectory Replanning for Quadrotors Using Kinodynamic Search and Elastic Optimization
Transferring Grasping Skills to Novel Instances by Latent Space Non-Rigid Registration
Translating Videos to Commands for Robotic Manipulation with Deep Recurrent Neural Networks
Ultra-Fast Multi-Scale Shape Estimation of Light Transport Matrix for Complex Light Reflection Objects
Uncertainty-Aware Learning from Demonstration Using Mixture Density Networks with Sampling-Free Variance Modeling
Unsupervised Learning of Hierarchical Models for Hand-Object Interactions
Using Information Invariants to Compare Swarm Algorithms and General Multi-Robot Algorithms
Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping.

« prev Page 7/8 next »