This is the repository that contains source code for the Project Instinct website.

Project Instinct explores instinct-level intelligences for whole-body control in robotics through deep learning approaches. Our research focuses on developing advanced control systems that enable robots to perform complex, dynamic tasks like parkour and agile locomotion.

• Whole-body Control: Coordinated control of entire robotic systems for dynamic tasks
• Deep Learning for Robotics: Neural network approaches for motor control and perception
• Parkour and Agile Locomotion: High-performance, acrobatic robot movement capabilities
• Real-time Inference: Onboard deployment of learned policies on robotic platforms

• Deep Whole-body Parkour: Research on neural whole-body control for parkour
• Hiking in the Wild: Research on scalable perceptive locomotion for humanoids

If you find our work useful for your research, please cite our relevant publications.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.