Catalog

Simulation-based reinforcement learning (RL) has benefited the field of robotics, helping manufacturers automate complex tasks. With generative AI, robotic simulation has accelerated the development and training of models. NVIDIA Isaac Lab enables RL with realistic physics and high fidelity virtual worlds to train and test robots. This builders’ session dives deep into NVIDIA’s Isaac Lab, AWS Batch, and Amazon EC2 instances powered by NVIDIA GPUs. Through hands-on exercises, learn distributed training for robots across multiple GPU nodes, and explore the crucial sim-to-real transfer, sharing insights and strategies for effectively translating simulation learnings to real-world robotic applications. You must bring your laptop to participate.