snape/RVO2

Optimal Reciprocal Collision Avoidance (C++)

What it solves

It addresses the problem of reciprocal collision avoidance for multiple independent mobile robots or agents. It allows these agents to move through a shared workspace without colliding with each other, all without requiring communication between the agents.

How it works

The library implements the Optimal Reciprocal Collision Avoidance (ORCA) formulation. It ensures collision-free motion by assigning half of the responsibility for avoiding a pairwise collision to each involved agent. The process of finding the optimal action for an agent is converted into a low-dimensional linear program, resulting in smooth motions. The C++98 implementation supports two-dimensional spaces and uses OpenMP to parallelize simulations across multiple processors for efficiency.

Who it’s for

It is designed for developers of robotics simulations and multi-agent systems who need a high-performance way to manage agent movement in dense or complex environments.

Highlights

  • Supports thousands of agents with computation times in the millisecond range.
  • No communication required between agents to avoid collisions.
  • Provides a simple API for specifying static obstacles, agents, and preferred velocities.
  • Parallelized using OpenMP for increased performance.

Related

  • Project
    snape/RVO2-CS

    A C# implementation of the Optimal Reciprocal Collision Avoidance (ORCA) algorithm for enabling multiple independent agents to avoid collisions without communication.

  • Project
    leggedrobotics/ocs2

    A C++ toolbox for solving nonlinear optimal control problems, enabling real-time Model Predictive Control (MPC) for robotic systems with limited onboard computation.

  • Project
    Simple-Robotics/aligator

    Aligator is a C++/Python library for fast, constrained trajectory optimization in robotics. It offers a modelling API, efficient solvers (ProxDDP, FeasibleDDP), Pinocchio integration, optional Crocoddyl compatibility, and parallel CPU optimisations. Install via conda or build from source (CMake/Pixi). Use it to generate motion plans or run model‑predictive control on complex robots.

  • Project
    onnx/optimizer

    A C++ library for performing arbitrary optimizations on ONNX models, providing shared optimization passes to be reused across different backend implementations.

  • Project
    stack-of-tasks/pinocchio

    Pinocchio is a high‑performance C++ library (with Python bindings) for rigid‑body dynamics, offering analytical derivatives, contact solvers, closed‑loop handling, and AD support. It powers many robotics and simulation tools and can be installed via Conda or pip.