深慢Shimmer 深慢Shimmer arXiv:2606.26154v1 Announce Type: new Abstract: Autonomous microrobots navigating biological vasculature could enable targeted drug delivery and thrombolysis, yet training control policies for realistic environments remains an open challenge. Prior reinforcement learning (RL) studies of microrobotic navigation have been limited to idealized geometries that omit complex hydrodynamic flow fields, confined branching structures, and dense cellular obstacles found in vivo. Here, we develop a physical
This research represents a significant leap from idealized simulations to realistic biological environments, offering high technical value for autonomous medical intervention.