Combining learned data abstraction, nonlinear filtering, and time-optimal trajectory planning, the autonomous drone racing system achieved second place at the 2019 AlphaPilot Challenge, reaching speeds up to 8 m/s.

Philipp Foehn, Dario Brescianini, Elia Kaufmann et al.

Proposed RadarSLAM leverages radar geometry and probabilistic point clouds for robust large-scale mapping in all weather conditions, achieving 2.1854% translation error.

Ziyang Hong, Yvan Petillot, Sen Wang

TartanAir provides a large-scale, multi-modal synthetic dataset with challenging environments to advance visual SLAM robustness.

Wenshan Wang, Delong Zhu, Xiangwei Wang et al.

Using ν-Gap based dynamics similarity and Bayesian optimization, the proposed method efficiently selects source experiences, improving target robot performance by 62%.

Michael J. Sorocky, Siqi Zhou, Angela P. Schoellig

Proposes a CVaR-based submodular optimization algorithm with theoretical guarantees for risk-aware multi-robot coordination.

Lifeng Zhou, Pratap Tokekar

GTI leverages trajectory intersections with GMM and cVAE to enable robots to generalize long-horizon tasks from few demonstrations, outperforming BC and GCBC.

Ajay Mandlekar, Danfei Xu, Roberto Martín-Martín et al.

Proposes an adaptive task allocation framework using real-time performance feedback to update robot capability parameters without environmental models.

Yousef Emam, Siddharth Mayya, Gennaro Notomista et al.

Using MAP-Elites and CPPNs, EGAD generates 2000+ diverse objects for robotic grasp evaluation.

Douglas Morrison, Peter Corke, Jürgen Leitner

Navigation2 employs behavior trees on ROS2, achieving 37.4 miles of collision-free autonomous campus navigation in dynamic environments.

Steve Macenski, Francisco Martín, Ruffin White et al.

OURS adapts NetVLAD to polar radar, achieving F1=0.69 under random rotation versus 0.23 for VGG-16/NetVLAD.

Ştefan Săftescu, Matthew Gadd, Daniele De Martini et al.

Proposes a GNN-based decentralized multi-robot path planning framework, combining CNN feature extraction with imitation learning, achieving near-optimal success rates in complex environments.

Qingbiao Li, Fernando Gama, Alejandro Ribeiro et al.

AVID learns multi-stage tasks via CycleGAN pixel-level translation of human videos, requiring only 20 minutes of demonstrations.

Laura Smith, Nikita Dhawan, Marvin Zhang et al.

Proposes LiDAR Iris, a global descriptor combining LoG-Gabor filtering and Fourier transform for fast, rotation-invariant loop closure detection.

Ying Wang, Zezhou Sun, Cheng-Zhong Xu et al.

RoboNet dataset combined with visual foresight and inverse models enables cross-robot generalization, surpassing single-robot training with 4-20x less data.

Sudeep Dasari, Frederik Ebert, Stephen Tian et al.

TagSLAM integrates AprilTags with factor graph optimization, enabling robust monocular SLAM with high accuracy and real-time performance.

Bernd Pfrommer, Kostas Daniilidis

Proposes a RRT*-based online informative path planning algorithm, enhancing environment exploration and 3D reconstruction, outperforming existing methods.

Lukas Schmid, Michael Pantic, Raghav Khanna et al.

Utilizing self-supervised dense visual correspondence training to enhance visuomotor policy learning with only 50 demonstrations.

Peter Florence, Lucas Manuelli, Russ Tedrake

Oxford Radar RobotCar Dataset uses millimeter-wave FMCW radar for robust urban scene understanding, covering 280km in diverse weather conditions.

Dan Barnes, Matthew Gadd, Paul Murcutt et al.

Hierarchical sim2real framework enables zero-shot transfer of multi-agent manipulation policies trained in simulation, achieving over 80% success in real-world tasks.

Ofir Nachum, Michael Ahn, Hugo Ponte et al.

DoorGym uses domain randomization to train door-opening policies, achieving up to 95% success.

Yusuke Urakami, Alec Hodgkinson, Casey Carlin et al.