Arsalan Mousavian, senior robotics research scientist at NVIDIA, discusses robotics research and manufacturing applications with Norman at the Innovation Forum.
Mousavian was drawn to robotics because it connects software to action in the physical world. RoboCup, a Zoox internship on 3D detection for autonomous vehicles, and work at Google Brain Robotics on navigating homes helped him decide to pursue robotics research.
Mousavian describes NVIDIA as an AI platform company that builds hardware alongside software and tools. Its robotics work focuses on manipulation in kitchens and manufacturing, including object grasping, scene segmentation, and generating collision-free motions while a robot works.
Traditional systems estimate object poses and depend on 3D models, which can make them hard to scale and sensitive to errors. Mousavian says research is moving toward predicting goal positions directly, and toward models that can estimate the pose of new objects from a few images instead of requiring a CAD model or retraining for every part.
Manufacturing requires reliable systems that can recover from failures and handle variation such as changing light, stains, and part defects. NVIDIA generates varied training data in simulation, then uses deployment feedback to improve models; Mousavian says simulated models may begin at 80 or 90 percent performance. He identifies usability and programming costs as barriers, and describes partnerships, public code and models, and Isaac ROS as routes to deployment.
Mousavian points to aging populations, workforce shortages, and efforts to localize supply chains as reasons robotics adoption may grow. He envisions uses such as round-the-clock restaurant kitchens and care for older people at home. He also urges manufacturers to share difficult site conditions and open problems so research can address practical needs.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Arsalan Mousavian, senior robotics research scientist at NVIDIA, explains simulation-trained robotics and AI at the August Innovation Forum with host Norman and audience members.
NVIDIA founder and CEO Jensen Huang presents the company’s AI strategy and product announcements at GTC Berlin, with European partners and customers.
NVIDIA CEO Jensen Huang discusses Europe’s AI economy, NVIDIA’s AI infrastructure and agent platforms, and physical AI at GTC Berlin, with examples from European industry and NVIDIA partners.
The host and main speaker from Robotics Central explains Kawasaki’s staged plan to deploy Kaleido humanoid robots, and how Japan’s AI funding, Nvidia computing, and human remote operators fit into that effort.
Amulia Vishban, founder of Techable Ventures, discusses investing in physical AI with Dominic Madori Davis on TechCrunch’s Equity podcast.
Les Carpus, global head of physical AI for NVIDIA’s Inception program, discusses robotics startups and the path to a ChatGPT moment with TechCrunch reporter Tim at Disrupt 2026.
