Project Grant 2209796

This three-year, $199,872 project grant from the National Science Foundation's Computer and Information Science and Engineering program aims to augment the Chrono computer simulation platform to better simulate autonomous agents and human-autonomous agent interactions. Led by Carnegie Mellon University, the multi-disciplinary team of 40 researchers will upgrade Chrono to incorporate recent computational dynamics innovations to generate machine learning training data, facilitate comparison of competing autonomous system designs, and gauge design robustness through simulation of edge cases. Specifically, the team will integrate a faster differential algebraic equations solver, new contact friction problem solving approaches, real-time flexible-body dynamics simulation, terrain dynamics simulation capabilities, just-in-time compiling for hardware-optimized executables, mixed data representations for efficient state information storage, and a scalable multi-agent framework enabling distributed, real-time human-autonomous agent interaction simulation over the internet. Upon completion, Chrono will empower broader robotics, automotive, and other research communities.