ICRA 2017
307 papers
The 2017 International Conference on Robotics and Automation (ICRA) featured research centered on autonomous navigation, control systems, and human-robot collaboration. Research in aerial robotics included the development of estimation and planning frameworks for aggressive quadrotor flight using single cameras and inertial measurement units. Other studies addressed industrial and field robotics, such as planning and control for autonomous excavation.
In the domain of localization, research explored entropy-weighted particle filters that utilize vertical and road intensity information for vehicle positioning. Human-robot interaction was represented by methods designed to manage interaction wrenches during joint manipulation tasks. Furthermore, the conference included work on imitation learning within the framework of Riemannian manifolds. Prolific authors contributing to these themes included Vijay Kumar, Chris Brunner, and Gabriel Hottiger. These contributions highlight a focus on improving the sensing and operational capabilities of robots in dynamic or specialized environments.
Top Research Directions
Research directions most represented by ICRA 2017 papers.
Top Authors
Impactful Papers
Papers from ICRA 2017 that shaped research clusters.
Foundational
RoboFDM: A robotic system for support-free fabrication using FDM
foundational
Support-Free Fabrication via Multi-Axis
Vote3Deep: Fast object detection in 3D point clouds using efficient convolutional neural networks
foundational
Volumetric Voxel Grids for 3D Spatial Re
SemanticFusion: Dense 3D semantic mapping with convolutional neural networks.
foundational
Spatio-Temporal Bayesian Fusion for 3D S
DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
foundational
Direct 6-DoF Pose Regression
Combining self-supervised learning and imitation for vision-based rope manipulation
foundational
Action Discretization for Autoregressive
SemanticFusion: Dense 3D semantic mapping with convolutional neural networks.
foundational
Lifting 2D Features into 3D Spatial Repr
Influential
Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge
influential
Learning Kinematics and Intuitive Physic
Vote3Deep: Fast object detection in 3D point clouds using efficient convolutional neural networks
influential
Vertical Feature Collapse in Spatial Pro
Virtual vs. Real: Trading Off Simulations and Physical Experiments in Reinforcement Learning with Ba
influential
Cost-Normalized Acquisition for Multi-Fi
Multi-view self-supervised deep learning for 6D pose estimation in the Amazon Picking Challenge
influential
Automated Supervision for Scalable Robot
Leveraging the urban soundscape: Auditory perception for smart vehicles
influential
Overcoming Sensor Occlusion Through Coll
DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
influential
Spatiotemporal Feature Preservation with
Sequential Bayesian optimization as a POMDP for environment monitoring with UAVs
influential
Look-Ahead Planning via Online Tree Sear
Combining self-supervised learning and imitation for vision-based rope manipulation
influential
Learning Representations for Deformable
Joint pose and principal curvature refinement using quadrics
influential
Exploiting Sparsity in Second-Order Opti
Safe certificate-based maneuvers for teams of quadrotors using differential flatness
influential
Control Barrier Functions for Safe Auton
Papers
« prev Page 7/7