Meta's AI Models Support $41.5M RAMMP Project for Assistive Robotics

Meta's open-source AI vision models, DINO and Segment Anything Model (SAM), are being utilized in the $41.5 million Robotic Assistive Mobility and Manipulation Platform (RAMMP) project at the University of Pittsburgh's Human Engineering Research Laboratories (HERL). This project, funded by the Advanced Research Projects Agency for Health (ARPA-H), aims to improve the lives of the estimated 5.5 million wheelchair users in the United States, addressing the approximately 100,000 wheelchair-related injuries treated in US emergency departments each year.

The RAMMP consortium includes partners such as Kinova Robotics, LUCI Mobility, and several academic institutions, including Carnegie Mellon, Cornell, Northeastern, and Purdue. Mansoor Khan, an ARPA-H Program Manager, stated that the initiative seeks to enable Americans with limited mobility to live more independently and engage in work and leisure activities through advanced robotics.

Alicia Jackson, Ph.D., ARPA-H Director, emphasized the need for technology that adapts to users' needs rather than requiring them to adapt to the technology. Sivashankar Sivakanthan noted that performance in assistive robotics is not solely about benchmark accuracy but about reliability in everyday life scenarios. This project follows Meta's ongoing commitment to enhance assistive technology, as seen in previous initiatives like the Luna smart glasses and AI agents for WhatsApp business management.