David Vissière

The Navigation and Control Technology Inside the AR.Drone Micro UAV

Abstract This paper exposes the Navigation and Control technology embedded in a recently commercialized micro Unmanned Aerial Vehicle (UAV), the AR.Drone, which cost and performance are unprecedented among any commercial product for mass markets. The system relies on state-of-the-art indoor navigation systems combining low-cost inertial sensors, computer vision techniques, sonar, and accounting for aerodynamics models.

Iterative calibration method for inertial and magnetic sensors

We address the problem of three-axis sensor calibration. Our focus is on magnetometers. Usual errors (misalignment, non-orthogonality, scale factors, biases) are accounted for. We consider a method where no specific calibration hardware is required. We solely use the fact that the norm of the sensed field must remain constant irrespective of the sensors orientation. The proposed algorithm is iterative. Its convergence is studied. Experiments conducted with MEMS sensors (magnetometers) stress the relevance of the approach.

Experimental autonomous flight of a small-scaled helicopter using accurate dynamics model and low-cost sensors

Abstract In this paper, we address the problem of guidance and control of a small-scaled helicopter (Benzin Acrobatic from Vario™ with a 1.8 m diameter rotor) equipped with only low-cost sensors. These sensors are an Inertial Measurement Unit (IMU), a GPS, and a barometer, which represent a total cost of USD 3000 including one PC board and one micro-controller). By contrast to other experiments reported in the literature, we do not rely on any accurate IMU or GPS systems which costs are, separately, largely above the mentioned amount of USD 3000. To compensate the weaknesses of this low cost equipment, we put our efforts in obtaining an accurate flight dynamics model. This improves the prediction capabilities of our embedded Kalman filter that serves for data fusion. The main contribution of this paper is to detail, at the light of a successful reported autonomous hovering flight, the derivation of the model. We give numerous details about implementation and discuss the relevance of some modelling hypothesis based on our experience.

Using distributed magnetometers to increase IMU-based velocity estimation into perturbed area

We address the problem of position estimation for a rigid body using an inertial measurement unit (IMU) and a set of spatially distributed magnetometers. We take advantage of the magnetic field disturbances usually observed indoors. This is particularly relevant when GPS is unavailable (e.g. during military operations in urban areas). We illustrate our technique with several experimental results obtained with a Kalman filter. We also present our testing bench which consists of low cost sensors (IMU and magnetometers).

Experiments of trajectory generation and obstacle avoidance for a UGV

In this paper, we report results of investigations conducted on a mobile robotic experiment developed at the DGA Laboratoire de Recherche en Balistique et Aerodynamique in a joint work with Ecole des Mines. The vehicles under consideration are in fact similar to unicycles. We investigate a flatness based approach (combining open-loop optimization and closed loop tracking) and gyroscopic forces control laws. Experimental results are exposed. A theoretical proof of obstacle avoidance for a gyroscopic scheme is presented.

On-the-field calibration of an array of sensors

We address the problem of the calibration of an array of sensors by investigating theoretically and experimentally the case of 2 three-axis sensors. Our focus is on magnetometers that can be used in a low-cost inertial navigation system. Usual errors (misalignments, non-orthogonality, scale factors, biases) are accounted for. The proposed calibration method requires no specific calibration hardware. Instead, we solely use the fact that, if the sensor is properly calibrated, the norm of the sensed field must remain constant irrespective of the sensors orientation. Several strategies of calibration for an array of sensors are described along with the impact of (unavoidable) field disturbances. Experiments conducted with a couple of magneto-resistive magnetometers and data fusion results illustrate the relevance of the approach.

A Movement Monitor Based on Magneto-Inertial Sensors for Non-Ambulant Patients with Duchenne Muscular Dystrophy: A Pilot Study in Controlled Environment

Measurement of muscle strength and activity of upper limbs of non-ambulant patients with neuromuscular diseases is a major challenge. ActiMyo® is an innovative device that uses magneto-inertial sensors to record angular velocities and linear accelerations that can be used over long periods of time in the home environment. The device was designed to insure long-term stability and good signal to noise ratio, even for very weak movements. In order to determine relevant and pertinent clinical variables with potential for use as outcome measures in clinical trials or to guide therapy decisions, we performed a pilot study in non-ambulant neuromuscular patients. We report here data from seven Duchenne Muscular Dystrophy (DMD) patients (mean age 18.5 ± 5.5 years) collected in a clinical setting. Patients were assessed while wearing the device during performance of validated tasks (MoviPlate, Box and Block test and Minnesota test) and tasks mimicking daily living. The ActiMyo® sensors were placed on the wrists during all the tests. Software designed for use with the device computed several variables to qualify and quantify muscular activity in the non-ambulant subjects. Four variables representative of upper limb activity were studied: the rotation rate, the ratio of the vertical component in the overall acceleration, the hand elevation rate, and an estimate of the power of the upper limb. The correlations between clinical data and physical activity and the ActiMyo® movement parameters were analyzed. The mean of the rotation rate and mean of the elevation rate appeared promising since these variables had the best reliability scores and correlations with task scores. Parameters could be computed even in a patient with a Brooke functional score of 6. The variables chosen are good candidates as potential outcome measures in non-ambulant patients with Duchenne Muscular Dystrophy and use of the ActiMyo® is currently being explored in home environment. Trial Registration: ClinicalTrials.gov NCT01611597

S.P.8 Validation of linear accelerations and angular velocities to estimate the efficacy of a subject when performing a quantified task in a controlled environment

Abstract The need for measures specifically designed to assess the muscle strength and activity of upper limbs of non-ambulatory patients with neuromuscular diseases is a major challenge. Aiming for a regular monitoring of the natural history of these patients, and trying to provide a powerful system for quantitative measurements, a movement Holter monitor is being developed in France at the Institute of Myology in collaboration with SYSNAV Company. The device consists of a watch equipped with sensors and specific software. The measuring principle is based on the use of Micro-Electro-Mechanical Systems inertial sensors and magnetometers operated through magneto-inertial equations. The issue lies in determining reliable variables representing the physical muscle level and in quantification of movement of non-ambulatory patients from the signals recorded by sensors: angular velocity, magnetic field and linear acceleration in 3D. In a preliminary study, we used a wired prototype to record healthy subjects performing Box and Block test. For each task, the exponential regression between the obtained task score and the activity levels as estimated by different computing of recorded linear acceleration and angular velocity indicate a strong correlation ( R 2 from 0.95 to 0.97). The results from other tests like Minnesota (score), computer typing (number of characters) and number of a standardized movement mimicking spoon elevation to mouth similarly showed strong correlation R 2 ranging from 0.88 to 0.96. These data demonstrate that variables issued from the computing of linear acceleration and angular velocities may predict the efficacy of a subject in a quantified task performed in a controlled environment. We therefore produce a wireless prototype that is now being evaluated in non-ambulatory patients at home.

Timestamping for an array of low-cost sensors.

This paper describes a new industrial communicatio n protocol, dNet protocol suite, and its application in control systems. Protocols in this s uite have some useful properties known from higher communication protocols. For example, abstract of a ddresses and connection, encryption and application services. They also have some characteristics typic al rather for industrial protocols, e.g. simple implementation in microcontrollers, compatibility w ith existing bus standards (CAN), support for program downloading and device configuration. The s uite can be operated almost over any data link layer. Document then presents a typical way of prot ocol usage, definition of application services and networks management.Abstract This paper describes a technique for timestamping data from an array of sensors. This is a challenging issue that needs to be solved for the magnetometry-based navigation system presented here. This system uses finite difference schemes to evaluate spatial and time derivatives of a vector field. The accuracy of the timestamping method is critical. In the context of this particular application, we detail the various difficulties that must be circumvented. We present a software architecture, which was run on an experimental device.