Fig. 6 | Autonomous microrobot navigation upstream in a flow environment.

a. Schematic of the reward function adjustment to promote microrobot navigation close to the wall, minimizing drag. b. Graph showing reward progression over time for microrobots in normal (blue line) and stronger (green line) flow conditions, highlighting differences in learning and adaptation. In normal flow, rewards steadily improve and stabilize around 200,000 steps. In stronger flow, initial difficulties lead to more negative rewards, but the algorithm shows significant improvement by 400,000 steps. c. A schematic illustrates the behavior of the microrobot attached to the wall to avoid drag and move against the flow. d. An image sequence shows a microrobot navigating within a microfluidic channel under flow conditions. Initially, the microrobot encounters maximum drag, i.e., when it is at the center of the channel (0.0 s to 6.7 s) before moving towards the wall (7.8 s to 31.4 s), where the drag is minimal. The reduced drag forces at the wall facilitate more stable and controlled navigation. The red arrows indicate the direction of the fluid flow.