Akiko Mizutani

Biomimetic Attitude and Orientation Sensors

We developed and flight-tested two biomimetic sensors that use the spectral, spatial and polarization distribution of light in the environment for navigation and stabilization. A sky polarization compass was constructed and methodologies for precise calibration were developed. In static and flight testing, the calibrated device was found to be comparable in accuracy to a solid state magnetic compass. A biomimetic version of the optical stabilization organ of dragonflies known as the ocelli was constructed. A technique of spectral opponency in ultraviolet and green wavelengths was demonstrated to be effective in reducing the biasing effect of the sun. In flight testing, the biomimetic ocelli were implemented as part of the autopilot for maintaining level flight and shown to be effective. The successful results indicate that biomimetic sensors may have a role in the quest to miniaturize the autopilots of small unmanned aerial vehicles.

Bioinspired optical sensors for unmanned aerial systems

Insects are dependant on the spatial, spectral and temporal distributions of light in the environment for flight control and navigation. This paper reports on flight trials of implementations of insect inspired behaviors on unmanned aerial vehicles. Optical flow methods for maintaining a constant height above ground and a constant course have been demonstrated to provide navigation capabilities that are impossible using conventional avionics sensors. Precision control of height above ground and ground course were achieved over long distances. Other vision based techniques demonstrated include a biomimetic stabilization sensor that uses the ultraviolet and green bands of the spectrum, and a sky polarization compass. Both of these sensors were tested over long trajectories in different directions, in each case showing performance similar to low cost inertial heading and attitude systems. The behaviors demonstrate some of the core functionality found in the lower levels of the sensorimotor system of flying insects and shows promise for more integrated solutions in the future.

Integration and flight test of a biomimetic heading sensor

We report on the first successful development and implementation of an automatic polarisation compass as the primary heading sensor for a UAV. Polarisation compassing is the primary navigation sense of many flying and walking insects, including bees, ants and crickets. Manually operated polarisation astrolabes were fitted in some passenger airliners prior to the implementation of the global positioning system, to compensate for the overal degradation of magnetic and gyrocompass sensors in polar regions. The device we developed demonstrated accurate determination of the direction of the Sun, with repeatability of better than 0.2 degrees. These figures are comparable to any solid state magnetic compass, including flux gate based devices. Flight trials were undertaken in which the output of the polarimeter was the only heading reference used by the aircraft as it flew through GPS waypoints.

An algorithm for terrain avoidance using optical flow

A technique for terrain following using optical flow is presented. The technique is inspired by the distribution of facets in the compound eyes of many flying insects. The image interpolation algorithm is used to compute accurate optical flow values and subsequently, height above ground. Some novel techniques to allow the camera to be pointed in arbitrary directions rather than simply downwards were implemented. The algorithm was tested extensively in simulation using a simple control law based on the optical flow system updating the set point of a barometric altitude driven control loop. Ongoing field trials are producing promising results

The Dragonfly Flight Envelope and its Application to Micro UAV Research and Development

Abstract In this paper we present quantitative analysis of three dimensional trajectories of dragonflies under free flight conditions. The trajectories were captured while male insects were engaged in their normal behaviour of combat to protect oviposition sites along a stream. With measurements of speed, acceleration and turn rate of large dragonflies we have the means by which comparative studies can be done against other species and in different environments. Using physical scaling laws we propose a method by which this data set can be used to provide a comparison for larger flapping wing UAV concepts. Our ultimate goal is to provide a robust standard against which flapping wing aircraft performance can be compared so that appropriate evolutionary pressure can also be applied to technological developments, thus freeing resources for truly viable designs.

Vertically displaced optical flow sensors to control the landing of a UAV

We report on an optical device to aid landing of Unmanned Aerial Systems. Optical flow calculates a measure of observed angular movement of the image seen by the sensor. Optical flow can be converted to range if speed of the sensor is known. Our approach eliminates this requirement by using two optical flow sensors displaced vertically to calculate range. The initial implementation was tested on an instrumented UAV with promising results. We show that the sensor provides useful range measurements at a height of several meters. We argue that this technique is comparable to vision techniques such as stereo in this application. We show alternate implementations with optics that do not require vertical displacement.

Semiautomatic calibration and alignment of a low cost, 9 sensor inertial magnetic measurement sensor

As UAVs and sensor networks become ubiquitous, cost and accuracy will be increasingly traded. We have developed techniques for calibration that consider this problem. Our technique performs a nonlinear optimization through all variables associated with zeroth and first order affects on accuracy of the sensors. The optimization is constrained by the known properties of the magnetic and gravitational fields. A miniature autopilot is presented as an example of the pressing need for automation of calibration.

Flight Test of a Polarisation Compass Integrated into an Unmanned Aerial Vehicle Autopilot

Abstract We report on the first successful development and implementation of an automatic polarisation compass as the primary heading sensor for a UAV. Polarisation compassing is the primary navigation sense of many flying and walking insects, including bees, ants and crickets. Manually operated polarisation astrolabes were fitted in passenger airliners operating over the arctic prior to the implementation of the global positioning system, to compensate for the overall degradation of navigation sensors in polar regions. The device we developed demonstrated accurate determination of the direction of the Sun, with repeatability of better than 0.2 degrees. These figures are comparable to any solid state magnetic compass, including flux gate based devices. Flight trials were undertaken in which the output of the polarimeter was the only heading reference used by the aircraft as it flew GPS waypoints and fixed heading commands without GPS.

Control of unmanned aerial vehicles using insect optical sensors

The key principles of insect vision have been deduced over decades of biological research.1, 2 A theme that has emerged is the reliance of insects on the spatial and temporal distribution of light for flight control. Computational foundations and control laws derived from insects3 have become increasingly well formulated. In the same period, unmanned aerial vehicles (UAVs) have emerged as a revolution in air power. However, UAVs are not as autonomous as we would like. Unlike insects, they depend on artificial sources for position information and do not operate well close to obstacles. As we attempt their miniaturization, scaling problems emerge beyond the reduced fidelity of small sensors. For example, the vibration of flapping wings degrades inertial sensing, while densely packed avionics results in distorted measurements of the earth’s magnetic field. Insects have overcome these problems using a highly integrated visuomotor system. They use the polarization pattern in the sky for heading, stabilize using the horizon as a visual reference, and navigate over terrain using vision. Many of the reflexes of insects only function as part of an ensemble of reflexes, analogous to the biologically inspired subsumption architecture proposed by Brooks.4 The first task of any UAV is stabilization. Agile airframes must be actively stabilized by their flight-control systems. We have reverse engineered the basic function of ocelli—an auxiliary visual system found in most insects—and tested their performance as part of the flight-control system of the UAV shown in Figure 2. Ocelli are composed of simple eyes, as distinct from compound eyes (see Figure 1). The ocelli of dragonflies have wide fields of view and a retina containing UV and green photoreceptors. They are anatomically adapted for sensing attitude and provide Figure 1. Head of a dragonfly, showing the dorsal-rim area and the lateral and median ocelli in relation to the compound eye.

Dragonfly hover is primarily mediated by vision

The sensory means by which hover is achieved could be inertial, visual or an unexplained sensory modality. Dragonflies in their natural habitat were shown not to maintain a stationary position in wind. Their position fluctuated significantly while returning to the original position. The movement of the dragonfly is correlated with the movement of vertically standing vegetation. This response would be non-causal with wind for an inertial or putative pressure based internal sensory system. It is postulated that with a substrate of moving water, sensitivity to movement on the visual horizon for controlling hover is a robust strategy.