Nearly two decades after introducing a widely used approach to collision avoidance, a team of University of Maryland researchers has been recognized with a top honor from the IEEE Robotics and Automation Society.
A paper co-authored by Dinesh Manocha and Ming Lin, "Reciprocal Velocity Obstacles for Real-Time Multi-Agent Navigation," has been named the Most Influential Paper from the IEEE International Conference on Robotics and Automation (ICRA) for the 2004–2008 period. The award will be presented at the conference in Vienna, Austria, in June.
The recognition highlights the long-term impact of the work, selected from more than 3,800 papers published at ICRA during those years. The award is given to research that has demonstrated sustained influence within robotics and across related fields.
Manocha and Lin, both Distinguished University Professors of computer science, co-authored the paper with their then-postdoctoral researcher Jur van den Berg.
Manocha has an appointment in the University of Maryland Institute for Advanced Computer Studies (UMIACS) and both he and Lin are active in the University of Maryland Center for Machine Learning.
Their paper introduced a collision-avoidance algorithm known as Reciprocal Velocity Obstacles (RVO), which models how moving agents—such as robots or people—adjust their paths while anticipating that others will do the same. By combining geometric reasoning with optimization-based velocity selection, the method allows agents to navigate shared spaces efficiently without direct communication.
Subsequent work by their students expanded the framework into several widely used variants, including AVO, BRVO, GVO, HRVO and ORCA.
The researchers demonstrated that the approach produces stable, smooth motion by reducing the need for repeated course corrections. It can also scale to environments with thousands of moving agents and account for both static and dynamic obstacles.
“RVO ensures that a moving robot or agent can avoid collisions with its environment and other agents, including people, animals and other robots, helping improve safety for both the system and those nearby,” the researchers said.
Since its introduction at ICRA in 2008, the RVO framework has been adopted across academia and industry. Open-source implementations have been downloaded tens of thousands of times and integrated into robotics libraries such as ROS, as well as game engines, crowd simulation systems and animation tools.
—News brief adapted from an article by the Department of Computer and Electrical Engineering