ZJU-FAST-Lab/ego-planner-swarm

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2151 stars · 379 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

EGO-Swarm is a decentralized, asynchronous trajectory planner for single and multi-agent quadrotor navigation in unknown, obstacle-rich environments, running on onboard resources. It extends EGO-Planner to swarm navigation and is built as a ROS package in C++ with an included UAV simulator.

Use cases

• plan collision-free trajectories for a swarm of drones in cluttered environments
• simulate multi-quadrotor autonomous navigation in unknown scenes
• run decentralized onboard motion planning for multicopters
• research and benchmark drone swarm navigation algorithms
• visualize planned drone trajectories in rviz

When to choose

• you need efficient local trajectory planning for one or many quadrotors
• you want a decentralized swarm navigation solution using only onboard compute
• you work in ROS/ROS2 with C++ and need a research-grade planner with a simulator

When to avoid

• you need planning for ground robots or non-aerial vehicles
• you require a global planner with prior maps rather than local replanning in unknown environments
• your project cannot use GPL-3.0 licensed code

Facets

library · maturity stable

simulation robotics machine-learning robotics autonomous-vehicles simulation cpp trajectory-planning drone-swarm quadrotor motion-planning ros obstacle-avoidance multi-agent linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ZJU-FAST-Lab/ego-planner-swarm")