Bio-inspired optical flow circuits for the visual guidance of micro air vehicles

Ruffier, F.; Viollet, S.; Amic, S.; Franceschini, N. · 2003 · OpenAlex-citations

DOI: 10.1109/iscas.2003.1205152

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of autonomous visual guidance for Micro Air Vehicles (MAVs), specifically focusing on altitude control and terrain following. The authors argue that conventional image processing systems are too heavy and energy-intensive for MAVs, which have payload constraints measured in grams. Instead, they propose a bio-inspired approach modeled after insect flight control, where altitude is regulated by maintaining a constant optic flow (retinal slip speed) rather than measuring absolute distance. The system, named OCTAVE (Optical altitude Control sysTem for Autonomous VEhicles), aims to demonstrate that a simple, low-complexity visual sensor can robustly guide a micro-robot over unknown terrain. To achieve this, the researchers developed a compact Elementary Motion Detector (EMD) circuit. They compared three implementations: an original analog SMD circuit, a Field Programmable Analog Array (FPAA) circuit, and a digital microcontroller-based circuit using a Cygnal 8051 F300 µC. The digital µC-based design proved superior, weighing only 0.8 grams and consuming 40 mW, compared to the bulkier and more power-hungry analog and FPAA alternatives. This circuit processes downward-facing visual input from two photodiodes to estimate angular velocity. The system was tested on a 100-gram tethered helicopter mounted on a pantographic whirling arm, allowing controlled horizontal movement over a textured arena with varying relief. The experimental results demonstrated that the OCTAVE system successfully enabled the MAV to follow terrain automatically. By servoing the rotor speed to maintain a reference optic flow, the helicopter adjusted its altitude in response to ground elevation changes. The system proved robust across a wide range of forward speeds (1 to 3 m/s) without requiring speed-specific trimming. Notably, the system automatically generated a "safe altitude" that increased with flight speed, providing greater clearance at higher velocities. The robot reliably avoided crashing despite changes in ground relief, even though it possessed no onboard knowledge of its actual altitude or speed values. The significance of this work lies in demonstrating that complex autonomous flight does not require heavy computational resources or prior maps of the environment. The OCTAVE system shows that a minimal visuo-motor control loop, inspired by insect biology, can provide robust terrain following with negligible weight and energy consumption. This approach offers a viable solution for MAVs, spacecraft, and submarines operating in uncertain environments, where payload limitations preclude traditional computer vision systems. The study validates the feasibility of mounting such lightweight EMD circuits onboard future micro-robots to close the control loop locally.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• perception action locomotion