Feimo-Yi Technology
An embodied intelligence startup founded by Liu Nianqiu, a former core member of DeepRoute.ai. The company focuses on transferring map-free autonomous driving paradigms to robotics applications.
Timeline (last 90 days)
Feimo-I Technology was founded by Liu Nianqiu, a veteran of DeepRoute.ai, entering the embodied intelligence sector.
Feimo-I Technology received a nearly hundred-million-dollar seed round led by IDG Capital, with participation from Yunqi Capital, DiDi, Fosun RZ Capital, Hesai Technology, and others.
Feimo-I adopts the term 'Physical Agent' rather than 'Physical AI,' positioning itself as an intelligent agent capable of assisting humans in completing tasks in the physical world, with the ability for cognitive-space decision-making, action-space planning, contact-space detail alignment, and memory.
Feimo-I targets open dynamic scenes with ambiguous boundaries (commercial districts, office buildings, residences, activity areas) rather than scenes with clear boundaries such as factory bolt-tightening or hotel item delivery, in pursuit of cross-scene transfer capability.
Feimo-I adheres to a data-driven approach, does not use the HD maps or extensive rule-writing of early autonomous driving, regards this as key to being AI Native, and believes that rules become a burden rather than an advantage as boundaries expand.
Feimo-I plans to release a hardware body this year and layer on a large model for a deployment demonstration, launch proprietary device data collection, begin multi-site deployment next year, and complete the 'deployment—data—training—re-deployment' closed loop.
All Facts (6)
Feimo-I Technology was founded by Liu Nianqiu, a veteran of DeepRoute.ai, entering the embodied intelligence sector.
50%UnverifiedFeimo-I adopts the term 'Physical Agent' rather than 'Physical AI,' positioning itself as an intelligent agent capable of assisting humans in completing tasks in the physical world, with the ability for cognitive-space decision-making, action-space planning, contact-space detail alignment, and memory.
50%UnverifiedFeimo-I targets open dynamic scenes with ambiguous boundaries (commercial districts, office buildings, residences, activity areas) rather than scenes with clear boundaries such as factory bolt-tightening or hotel item delivery, in pursuit of cross-scene transfer capability.
50%UnverifiedFeimo-I adheres to a data-driven approach, does not use the HD maps or extensive rule-writing of early autonomous driving, regards this as key to being AI Native, and believes that rules become a burden rather than an advantage as boundaries expand.
50%UnverifiedFeimo-I Technology received a nearly hundred-million-dollar seed round led by IDG Capital, with participation from Yunqi Capital, DiDi, Fosun RZ Capital, Hesai Technology, and others.
50%UnverifiedFeimo-I plans to release a hardware body this year and layer on a large model for a deployment demonstration, launch proprietary device data collection, begin multi-site deployment next year, and complete the 'deployment—data—training—re-deployment' closed loop.
50%