Closing the Sim-to-Real Gap with Unreal Engine

Explore how to bridge the simulation-to-reality gap in robotics and AI development using Unreal Engine's advanced capabilities in this 50-minute conference talk from Unreal Fest Orlando 2025. Learn from Jim Piavis of Scaled Foundations as he demonstrates how Open GRID leverages Unreal Engine as the foundation for cloud-based AI model development, incorporating AirGen (formerly AirSim) for physics simulation and integrating with extensive AI model catalogs. Discover how high-performance graphics, procedural rendering, and Nanite tessellation enable real-time, interactive simulation environments for robots and their sensors. Understand the process of adapting AI models including object detection, segmentation, and visual navigation to compose and integrate macro-scale tasks that respond to changing environments. Examine the perceived gap between simulation and real-world performance, explore the continuum of risk reduction techniques required to close this sim-to-real gap, and see how Scaled Foundations uses highly realistic virtual environments to demonstrate the deployment of robust and safe robotic intelligence behaviors across multiple robotic form factors from simulation to real-world applications.