FAVbot: An Autonomous Target Tracking Micro-Robot with Frequency Actuation Control

TL;DR

This work tackles the challenge of achieving autonomous, vision-guided steering in centimeter-scale micro-robots with minimal actuation hardware. It introduces FAVbot, a 3-cm robot that uses a single piezoelectric actuator whose resonance modes are controlled by actuation frequency to steer, paired with an on-board CNN-based vision system for closed-loop target tracking. The approach yields multiple resonant motion modes (straight, turn, and complex trajectories) within a compact footprint, enabling untethered operation and autonomous navigation. The system delivers up to 15 minutes of untethered operation and demonstrates autonomous tracking of static and moving targets, illustrating a significant step toward AI-enabled autonomy in constrained-scale robots. The work points to future improvements via thin-film materials and ASIC integration to further shrink size and enhance efficiency and multi-target capabilities.

Abstract

Robotic autonomy at centimeter scale requires compact and miniaturization-friendly actuation integrated with sensing and neural network processing assembly within a tiny form factor. Applications of such systems have witnessed significant advancements in recent years in fields such as healthcare, manufacturing, and post-disaster rescue. The system design at this scale puts stringent constraints on power consumption for both the sensory front-end and actuation back-end and the weight of the electronic assembly for robust operation. In this paper, we introduce FAVbot, the first autonomous mobile micro-robotic system integrated with a novel actuation mechanism and convolutional neural network (CNN) based computer vision - all integrated within a compact 3-cm form factor. The novel actuation mechanism utilizes mechanical resonance phenomenon to achieve frequency-controlled steering with a single piezoelectric actuator. Experimental results demonstrate the effectiveness of FAVbot's frequency-controlled actuation, which offers a diverse selection of resonance modes with different motion characteristics. The actuation system is complemented with the vision front-end where a camera along with a microcontroller supports object detection for closed-loop control and autonomous target tracking. This enables adaptive navigation in dynamic environments. This work contributes to the evolving landscape of neural network-enabled micro-robotic systems showing the smallest autonomous robot built using controllable multi-directional single-actuator mechanism.

Figures (10)

• Figure 1: FAVbot: 3-cm miniaturized robot. (a) Conceptual rendering of robot autonomous target tracking operation. (b) Image of the assembled robot. The diameter of the robot is 3 cm. (c) Sub-modules of the robot, including on-board power, CNN-based computer vision, and frequency-controlled single-actuator steering.
• Figure 2: (a) Circuit diagram. (b) Circuit components. COTS components has been used except the custom driver. The circuit component assembly (without the piezoelectric buzzer) occupies less than 23 $\times$ 26 $\times$ 28 mm$^3$ volume.
• Figure 3: Mechanical design of FAVbot.
• Figure 4: Resonance mode shapes from finite elements analysis.
• Figure 5: Frequency-controlled steering. (a) A selection of representative motion patterns of FAVbot under various frequencies between 1 - 75 kHz. Arrows plot the instantaneous orientations of the robot. Color represents the elapsed time. (b) Extracted average values of the linear, transnational, lateral, and angular speed of the robot under different frequencies. Labels a - w are consistent in both plots. Videos of some modes can be found in the multimedia attachment.