Robert Boschgmbh

Rudolf Bichler

Thomas Rocznik

Christian Peters

Gerhard Lammel

Rainer Dorsch

Akshay Malhotra

Markus Dihlmann

Robert Duerichen

Guido Retz

Amithash Kankanallu Jagadish

Kejia Wang

A system and method is disclosed for classifying time-series data provided to a machine-learning model from a continuous sensor signal. The data may be 'windowed' or “divided” into a smaller data segment using a first stage classifier where an “event of interest” may be identified. The first stage classifier may employ an algorithm that prohibits false negative identifications. The data segment detected as including an event of interest may then be transmitted to a second stage classifier operable to performs a full classification on the data segment. The multi-stage network may require less power and a less complex structure.

System and method for distributed neural networks on edge devices

A cascaded machine learning inference system and method is disclosed. The cascaded system and method may be designed to be employed in resource restricted environments. The cascaded system and method may be applicable for applications that operate with limited power (e.g., a wearable smart watch). The cascaded system and method may employ two or more subsystems that are operable to classify an input signal provided by any number or types of sensors suitable for a given application. For instance, the sensors used may include gyroscopes, accelerometers, magnetometers, or barometric altimeters. The system and method may also be further split functionality across additional or new subsystems. By splitting operations and functionality across additional subsystems, the overall power consumption may further be reduced.

State-aware cascaded machine learning system and method

A method for determining the spatial orientation of an object encompasses the steps of detecting a rotational speed of the object with the aid of a first sensor; determining an estimated value for the spatial orientation of the object based on the detected rotational speed and a starting value for the spatial orientation; detecting the acceleration aacting on the object with the aid of a second sensor; determining the acceleration component of the detected acceleration aresulting from the rotational movement, based on the detected rotational speed and a given distance of the second sensor from the rotational axis of the rotational movement; determining the acceleration component aof the detected acceleration aresulting from the gravity; and correcting the estimated value for the spatial orientation, taking into account the acceleration component aresulting from the gravity.

Determination of a spatial orientation

A sensor for acquiring measured values and for outputting data samples having at least one first register for storing a sensor time, which includes time information on phase position and/or period of the data samples, the first register being readable out externally. The sensor includes at least one second register, which is writable on externally, and by which the phase position and/or the period of the data samples is able to be set in the sensor. Also described is a sensor system having at least one sensor and an external control unit by which at least the first register is readable and at least the second register is writable. Also described is a sensor-fused system having at least one sensor system having at least two sensors. Also described is a method for setting the phase position and/or the period of data samples in a sensor system or a sensor-fused system.

Sensor time synchronization

Growing community of inventors

Reutlingen, Germany

Rudolf Bichler