GAINS: Game-Dev AI NPC Simulation

GAINS: This team uses Unreal Engine to develop AI-driven NPCs and interactive environments for use in simulating complex human behaviors and dynamics. Students will develop skills in NPC and environment/level scripting, perception, and interactivity through machine learning based NPC performance trials.

Advisors:

Description:

This team uses Unreal Engine to develop AI NPC simulations on interactive environment levels that mirror real-life human behaviors and environmental interactions. Students will develop NPCs with robust perception, movement, and action/reaction behavioral traits. Custom level designs will simulate real-world environments and locations for interaction with automated NPC actions. These efforts will allow architectural design and planning professionals to test designs to improve safety, access, programmatic use, and social equity in the built environment while fostering skills valued in video game development, safety and pedestrian/automotive testing, and other related simulation-based studies.

 

Activities/Technologies

Students will need to have a computer capable of running Unreal Engine 5+ and have local rights to install version control software (Perforce). Knowledge of and/or interest in C++ is a bonus. Attendance in weekly formal group meetings and informal sub-team meetings are required. Discussion and progress posting will occur via Discord. Opportunities for student leadership and sub-team management are encouraged.

 

Prerequisites:

The members are expected to have finished one semester of calculus and one programming course.

Publications:

Barbarash, D. 2022. Automated Recording of Human Movement Using an Artificial Intelligence Identification and Mapping System. Journal of Digital Landscape Architecture vol 7, Harvard University, Cambridge, Boston. 59-70. DOI: 10.14627/537724007

Barbarash, David, et al. "Artificial Intelligence Systems for Automated Site Analytics and Design Performance Evaluation."  Landscape Research Record, no. 9, March 2020, pp. 192-203.  https://thecela.org/wp-content/uploads/LRR_v.9_FINAL_2020_Reduced-1.pdf