Commissariat &#xc0; L'&#xc9;nergie Atomique Et Aux &#xc9;nergies Alternatives

Commissariat a L'energie Atomique

Movea

Movea Sa

Andréa Vassilev

Ghislain Despesse

Tristan Hautson

Timothée Jobert

Saifeddine Aloui

David Dominique

Roland Blanpain

Bruno Flament

Dominique David

Yanis Caritu

Jean-Philippe Gros

Sébastien Brulais

Thomas Jager

Leandro Cassarino

Cindy Bassompiere

Jean-Jacques Chaillout

A method and a device for multi-class classification of an application relative to a classification of a transport mode by machine learning, including: (a) sensors configured to measure at successive instants physical quantities specific to said application, and (b) a microprocessor configured to: (b1) acquire said successive measurements from the sensors to calculate predictors from physical quantities of said measurements, (b2) construct continuous series (a) of samples, each sample being a vector formed of a predetermined number of predictors, (b3) break down each series (a) of samples into a set of sequences (S) each corresponding to a specific class among a set of predetermined classes, (b4) under-sample the sequences associated with the classes among said set of predetermined classes while conserving for each selected sequence a predetermined fraction of samples of said sequence, thereby forming classes associated with balanced samples, and (b5) construct a classifier (F) from said balanced samples by machine learning.

Device and method for multi-class classification by machine learning

A system for detection of an air transport made includes a barometric sensor to measure the values of the atmospheric pressure, and a processing unit to acquire the pressure values and to identify if the transport mode is air transport (aircraft mode), depending on the pressure measurements.

Process and system for detecting an aerial transport mode

The method involves a step of measurement of a bearing force (Fp) on a pedal () of the pedal set (), different from the useful effort, exerted by the individual during a pedalling cycle, a step of determination of a maximum (max(Fp)) of the bearing force measured during the pedalling cycle (CP), and a step of estimation of a mean useful effort exerted by the individual during said pedalling cycle from the maximum of the bearing force determined and with the aid of a predetermined correlation function between a maximum bearing force during a pedalling cycle and a mean useful effort in said cycle.

Method and a system for estimation of a useful effort provided by an individual during a physical activity consisting in executing an alternating pedalling movement on a pedal device

Locating a moving magnetic object includes determining its position or orientation from measurements from an array N tri-axial magnetometers (N&#x3e;5) mechanically linked to one another with 0-DOF by repeatedly estimating its position relative to the array from measurements made either in a preceding iteration or from a sensor distinct from the array's magnetometers before making a new measurement using the array's magnetometers; computing the distance between each magnetometer and the object's estimated position; eliminating Ni magnetometers closest to this estimated position, where Ni&#x3c;N; using the remaining magnetometers, making a new measurement of the field generated or modified by the object; and determining its new position or orientation from the new measurements so as to obtain the object's new location.

Method and device for locating a magnetic object

A method of detecting a point of contact between a tip of an instrument and a writing support, the instrument being one of a pen and an eraser, and the method comprising determining, with the aid of a magnetic object fixed without any degree of freedom on the instrument, the height of a point of the instrument with respect to a bearing face of a tablet. The bearing face constituting the writing support or supporting the writing support. The method further includes: comparing the height determined with a preconfigured contact threshold, and detecting a point of contact between the tip of the instrument and the writing support if the height determined is less than the contact threshold and, in the converse case, detecting no point of contact between the tip of the instrument and the writing support.

Method of detecting point of contact between a tip of a utensil and writing support

The invention applies to a pointing device of a mobile element, for example a cursor, on a plane surface. The pointing device comprises a first sensor for measuring the angular rates of the device and a second sensor of the linear accelerations along three dimensions of said device. Preferably, the first sensor is a two-axis or three-axis gyrometer and the second sensor is a three-axis accelerometer. The invention makes it possible to render the movements of the mobile element in the surface of the orientation in which the pointing device is held by its user. This result is achieved by global resolution of the angles of torsion by combining the measurements of the first and second sensors either within an extended Kalman filter or by application of an optimization criterion. The invention also provides a method for estimating the parameters which characterize motion and/or orientation of an object in space from said combination of the measurements of the first and second sensors either within an extended Kalman filter or by application of an optimization criterion.

Motion capture pointer with data fusion

Sports ball, and more generally a pneumatic object with at least one internal pressure sensor () of said ball, automatic inflator (), capable of compensating for a decrease in the internal pressure, this inflator having at least one gas micro-generator, said at least one micro-generator being capable of being actuated when the pressure sensor () measures an internal pressure below a predetermined threshold.

Growing community of inventors

Grenoble, France

Andréa Vassilev