University of Maryland Has Strong Presence at ICRA 2024
University of Maryland (UMD) researchers will present 17 papers at the International Conference on Robotics and Automation (ICRA) in Japan this week, detailing innovative work navigation, decision-making, perception systems, reinforcement learning, and other topics of high interest to researchers in robotics and related fields. The ICRA, organized annually by the Institute of Electrical and Electronics Engineers (IEEE), is the world’s largest academic conference in robotics, with the highest impact factor of any robotics conferences or journal. UMD faculty presenting at the event include Professor Yiannis Aloimonos (computer science/electrical engineering), Professor Nikhil Chopra (mechanical engineering), Research Scientist Cornelia Fermuller (Institute for Advanced Computer Studies), Distinguished Professor Ming C. Lin (computer science), Distinguished University Professor Dinesh Manocha (computer science/electrical engineering/Institute for Systems Research), Assistant Professor Abhinav Shrivastava (computer science), Associate Professor Pratap Tokekar (electrical and computer engineering), and Professor Miao Yu (mechanical engineering). “Our presence at ICRA reflects the depth and breadth of robotics research at UMD, as well as our strong interdisciplinary focus,” said Manocha, who is one of the world’s leading experts on robotics and autonomy. “We’re able to bring together computer scientists, engineers, and colleagues from other disciplines to tackle some of the most challenging problems in the field.” UMD’s Maryland Robotics Center (MRC), launched in 2010, has been a key driver for such synergy. This unique research hub, which also sponsors educational and outreach programs, was designed specifically to foster collaboration among a wide range of faculty members with an interest in robotics. “We are extremely pleased at the level of participation by University of Maryland faculty, staff, and students at premier robotics conferences like ICRA, IROS, and RSS,” MRC Director Derek Paley (aerospace engineering) said. Specific topics to be covered by UMD researchers at ICRA 2023 include trajectory planning for outdoor environments, which can present difficulties due to uneven terrain. A paper by Lin, Manocha, and multi-institutional collaborators details a machine learning algorithm designed to plot environment-specific trajectories and minimize the risk of collisions. Aloimonos, Chopra, Tokekar, and Yu, meanwhile, will present UIVNav, an innovative navigation system for underwater robots, designed to steer them towards objects of interest while avoiding obstacles, Multiple papers reflect UMD’s ongoing collaboration with the U.S. Army Research lab under the five-year AI and Autonomy for Multi-Agent Systems (ArtIAMAS) program, led by Paley and Manocha, which aims to spur development of complex, autonomous systems that operate collectively across multiple domains. In addition, UMD researchers will present work conducted jointly with colleagues at Amazon 126, which has been supporting UMD robotics research through student fellowships and seed grants.
1. MTG: Mapless Trajectory Generator with Traversability Coverage for Outdoor Navigation 2. AG-CVG: Coverage Planning with a Mobile Recharging UGV and an Energy-Constrained UAV 3. AcTExplore: Active Tactile Exploration on Unknown Objects 4. Collaborative Decision-Making Using Spatiotemporal Graphs in Connected Autonomy 5. Pre-Trained Masked Image Model for Mobile Robot Navigation 6. UIVNAV: Underwater Information-Driven Vision-Based Navigation Via Imitation Learning 7. UAV-Sim: NeRF-Based Synthetic Data Generation for UAV-Based Perception 8. Task-Driven Domain-Agnostic Learning with Information Bottleneck for Autonomous Steering 9. Unconstrained Model Predictive Control for Robot Navigation under Uncertainty 10. VAPOR: Legged Robot Navigation in Unstructured Outdoor Environments Using Offline Reinforcement Learning 11. MIM: Indoor and Outdoor Navigation in Complex Environments Using Multi-Layer Intensity Maps 12. HandyPriors: Physically Consistent Perception of Hand-Object Interactions with Differentiable Priors 13. WayEx: Waypoint Exploration Using a Single Demonstration 14. Can an Embodied Agent Find Your “Cat-Shaped Mug”? LLM-Based Zero-Shot Object Navigation 15. Cook2LTL: Translating Cooking Recipes to LTL Formulae Using Large Language Models 16. Sim-To-Real Robotic Sketching Using Behavior Cloning and Reinforcement Learning 17. GrASPE: Graph based Multimodal Fusion for Robot Navigation in Outdoor Environments”
May 13, 2024 Prev Next |
ICRA 2024