/RVO2-Java

Optimal Reciprocal Collision Avoidance (Java)

Primary LanguageJavaApache License 2.0Apache-2.0

Optimal Reciprocal Collision Avoidance for Java

https://gamma.cs.unc.edu/RVO2/

DOI

We present a formal approach to reciprocal collision avoidance, where multiple independent mobile robots or agents need to avoid collisions with each other without communication among agents while moving in a common workspace. Our formulation, optimal reciprocal collision avoidance (ORCA), provides sufficient conditions for collision-free motion by letting each agent take half of the responsibility of avoiding pairwise collisions. Selecting the optimal action for each agent is reduced to solving a low-dimensional linear program, and we prove that the resulting motions are smooth. We test our optimal reciprocal collision avoidance approach on several dense and complex simulation scenarios workspaces involving thousands of agents, and compute collision-free actions for all of them in only a few milliseconds.

RVO2 Library Java is an open-source Java 8 implementation of our algorithm in two dimensions. It has a simple API for third-party applications. The user specifies static obstacles, agents, and the preferred velocities of the agents. The simulation is performed step-by-step via a simple call to the library. The simulation is fully accessible and manipulable during runtime.

Build Status

SPDX-FileCopyrightText: 2008 University of North Carolina at Chapel Hill
SPDX-License-Identifier: Apache-2.0

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

  https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Please send all bug reports to geom@cs.unc.edu.

The authors may be contacted via:

Jur van den Berg, Stephen J. Guy, Jamie Snape, Ming C. Lin, Dinesh Manocha
Dept. of Computer Science
201 S. Columbia St.
Frederick P. Brooks, Jr. Computer Science Bldg.
Chapel Hill, N.C. 27599-3175
United States of America