← Projects
active

Isaac Drone Racer 2

Modernized autonomous drone racing RL framework for Isaac Sim 5.1 and Isaac Lab 2.3.2. Supports vision-based gate tracking, semantic segmentation, swarm racing with multiple agents, and multi-objective reward functions for high-speed navigation.

Reinforcement Learning Isaac Sim Aerial Robotics Vision-Based Control Swarm Robotics Autonomous Racing
View on GitHub ↗

Overview

Isaac Drone Racer 2 is a modernized autonomous drone racing reinforcement learning framework built on NVIDIA Isaac Sim 5.1, Isaac Lab 2.3.2, and Python 3.11. It extends the original pipeline with vision-based perception, swarm capabilities, and a reorganized codebase ready for camera-driven RL research.

Key Contributions

  • Framework Modernization: Upgraded full stack to Isaac Sim 5.1, Isaac Lab 2.3.2, and Python 3.11 for compatibility with current NVIDIA tooling
  • Reorganized Simulation Framework: Restructured codebase for vision-based drone racing, gate tracking, and learning-based control experiments
  • Camera-Based RL Platform: Prepared infrastructure for onboard sensing using semantic segmentation and simulated camera observations
  • Swarm Racing: Supports multiple agents racing simultaneously with multi-objective reward functions covering gate navigation, speed, and collision avoidance