image: (A) Coil system. (B) Tank and helical microswimmer. (C) Real experimental image. (D) Helical microswimmer with detailed shape parameters. The red dashed line indicates the body axis of the helical microswimmer.
Credit: Jiangfan Yu, School of Science and Engineering, The Chinese University of Hong Kong.
Recent advances have yielded significant progress in actuation, navigation, and control of magnetic microrobots. Nevertheless, dynamic obstacle avoidance in 3D environments remains a critical challenge, often relying on computationally intensive path-planning methods that limit real-time performance. "Using a hierarchical radar system to enable high-frequency direction updates minimizes computational load while ensuring collision-free navigation," explained corresponding author Jiangfan Yu, a professor at The Chinese University of Hong Kong, Shenzhen. The radar framework comprises (a) a motion sphere for directional sampling, (b) a detection sphere for obstacle monitoring, and (c) a coarse-to-fine search algorithm to balance precision and efficiency. "This integrated control scheme paves the way for reliable microrobot deployment in dynamic biological environments, avoiding trial-and-error navigation strategies," stated the authors. Thus, they proposed a helical microswimmer system featuring radar-based navigation fused with a neural-fuzzy motion controller to achieve real-time obstacle evasion.
The control scheme leverages hierarchical optimization and adaptive weighting to handle complex 3D spaces. The coarse-to-fine search reduced computational load by 85% compared to conventional RRT* methods, enabling a 2.6 Hz update frequency for motion direction. "The three navigation modes—free movement, static obstacle avoidance, and dynamic evasion—autonomously switch based on radar inputs, optimizing path efficiency," noted lead author Yuezhen Liu. The RBF-ELM neural network further compensated for directional deviations caused by fluidic disturbances, maintaining trajectory tracking errors below 145 µm.
Experiments in glycerol solution (950 cP) validated the system’s capability to navigate among 8 static and 8 dynamic obstacles (1.3 mm diameter). The microswimmer reached targets with a 97% success rate, demonstrating adaptability to randomly moving obstacles at speeds up to 125 µm/s. "Circumferential reinforcement of the helical body enabled stable propulsion under rotating magnetic fields (3 mT, 7 Hz), achieving vertical speeds of 180 µm/s," detailed Liu. However, the system struggled with high-inertia obstacles and exhibited minor trajectory overshoot during abrupt directional changes. Mechanical interference between closely spaced obstacles also occasionally triggered conservative evasion paths. Future work will address scalability in larger workspaces and integration with photoacoustic imaging for in vivo applications.
Authors of the paper include Yuezhen Liu, Yibin Wang, Kaiwen Fang, Hui Chen, Guangjun Zeng, and Jiangfan Yu.
This work was financially supported by the National Key R&D Program of China under Project No. 2022YFA1207100, the Guangdong Basic and Applied Basic Research Foundation under Project No. 2023A1515012973, the Shenzhen Science and Technology Program under Project No. JCYJ20241202124015021, the Shenzhen Institute of Artificial Intelligence and Robotics for Society under Project No. BN00202312037-1D, the China Merchants Group funding under Project No. BN00202312037, and the Longang District Shenzhen’s “Ten Action Plan” under Project No. LGKCSDPT 2024002 and 2024003.
The paper “Radar-Based Control of a Helical Microswimmer in 3-Dimensional Space with Dynamic Obstacles” was published in the journal Cyborg and Bionic Systems on Jun 2, 2025, at DOI: 10.34133/cbsystems.0158.
Journal
Cyborg and Bionic Systems