Motion controllers for 3D characters with contact rich interactions
The goal of this project will be to develop automatic controllers for 3D characters involved in contact rich physical interactions. This includes activities such as climbing, wrestling, and manipulation of objects. The research will integrate data captured from real world sensors (motion capture) to learn parameters for PD and non-linear feedback controllers. The learned controllers will then be used to generate motion for 3D characters in a physics-based video game environment. Another objective will be to evaluate the learned controllers and attempt to determine which features of the real and simulated environments produce accurate and robust control.
Masters degree in computer sciences, software engineering or a related field
- Knowledge of video games, physics simulation, and 3D graphics
- Experience with motion capture technology is helpful
- Knowledge of machine learning and numerical optimization
- Strong programming skills in C++ or Python