PAICar: a prototype of an embodied neuromorphic intelligent robot platform
Feng S, Liu M K, Zhong J Y, et al
Sci China Inf Sci, 2026, 69(6): 164401
Embodied neuromorphic intelligence (ENI) integrates neuromorphic computing with embodied perception-action loops, simulating the brain’s operational principles. It equips agents with perception, decision-making, and action capabilities, leveraging brain-inspired architectures for efficient and flexible computation. ENI emphasizes multimodal sensory integration and interaction with environments through motor systems. At its core, neuromorphic computing utilizes a spiking neural network (SNN) to process information in a manner similar to biological brains. Moreover, neuromorphic chips replicate neural structures, offering low power consumption and high parallelism, enabling real-time decisionmaking in dynamic environments and driving the evolution of robotic intelligence.