EP-4739534-A1 - METHOD FOR RIDER-SPECIFIC AND/OR BICYCLE-SPECIFIC DETERMINATION OF AT LEAST ONE BICYCLE FUNCTION ALGORITHM

EP4739534A1EP 4739534 A1EP4739534 A1EP 4739534A1EP-4739534-A1

Abstract

The invention relates to a method for the rider-specific and/or bicycle-specific determination of at least one bicycle function algorithm, said method comprising the following method steps: - acquiring sensor data from a plurality of cyclists and/or bicycles; - acquiring metadata that can be assigned to the sensor data; - determining a rider type based on the sensor data and the metadata; - determining a rider-specific bicycle function algorithm based on the sensor data and metadata that can be assigned to the rider type; - providing open-loop or closed-loop control of the bicycle and/or a bicycle component based on the rider-specific bicycle function algorithm.

Inventors

Schnee, Jan
PLETINCKX, JO
Laqua, Annika

Assignees

Robert Bosch GmbH

Dates

Publication Date: 20260513
Application Date: 20240627

Claims (14)

1 . Method for driver-specific and/or bicycle-specific determination of at least one bicycle function algorithm, comprising the following method steps: - Collecting sensor data from a large number of cyclists and/or bicycles; - Collecting metadata that can be associated with the sensor data; - Determining a driver type based on the sensor data and metadata; - Determination of a rider-specific bicycle function algorithm based on the sensor data and metadata attributable to the rider type; - Controlling or regulating the bicycle and/or a bicycle component based on the rider-specific bicycle function algorithm.
2. Method according to claim 1, characterized in that the determination of the driver type is based on a cluster analysis.
3. Method according to claim 2, characterized in that the sensor data and the metadata are correlated before the cluster analysis.
4. Method according to one of claims 2 or 3, characterized in that the sensor data and the metadata are standardized before the cluster analysis.
5. Method according to one of claims 2 to 4, characterized in that the cluster analysis is carried out by means of a machine learning model.
6. Method according to one of the preceding claims, characterized in that the sensor data are recorded by a sensor arranged on the bicycle and/or on the rider.
7. Method according to one of the preceding claims, characterized in that the sensor data is designed as heart data, in particular as pulse information, as driver performance, as speed, as acceleration, as brake pressure, as cadence, as yaw rate, as roll rate, as steering angle and/or as torque.
8. Method according to one of the preceding claims, characterized in that the metadata is designed as a feeling of exertion, as a driving experience, as a driving comfort and/or as a driving style.
9. Method according to one of the preceding claims, characterized in that the metadata is designed as route information, as environmental information and/or as gradient information
10. Method according to one of the preceding claims, characterized in that the bicycle function algorithm is stored locally.
11. Method according to one of the preceding claims, characterized in that a driving experience is determined based on the bicycle function algorithm and by sensor data of a bicycle recorded during a ride, wherein the driving experience is displayed on a display unit (34) of the bicycle.
12. Method according to one of the preceding claims, characterized in that a determination of switching information is carried out based on the bicycle function algorithm and by sensor data of a bicycle recorded during a ride, wherein a switching unit of the bicycle is controlled based on the switching information.
13. Method according to one of the preceding claims, characterized in that a determination of braking information is carried out based on the bicycle function algorithm and by sensor data of a bicycle recorded during a ride, wherein a brake support unit of the bicycle is controlled based on the braking information.
14. Bicycle, in particular an electric bicycle (16), with a control unit for controlling or regulating the bicycle, with a sensor unit for recording sensor data, wherein the bicycle and/or an external device (100) connected to the bicycle is controlled based on the bicycle function algorithm according to claim 1.

Description

Description title Method for driver-specific and/or bicycle-specific determination of at least one bicycle function algorithm State of the art EP 3 531 071 A1 describes a method for supporting a shared driving experience among drivers of a plurality of mobile units. disclosure of the invention The invention relates to a method for driver-specific and/or bicycle-specific determination of at least one bicycle function algorithm, comprising the following method steps: - Collecting sensor data from a large number of cyclists and/or bicycles; - Collecting metadata that can be associated with the sensor data; - Determining a driver type based on the sensor data and metadata; - Determination of a rider-specific bicycle function algorithm based on the sensor data and metadata attributable to the rider type; - Control or regulation of the bicycle and/or a bicycle component based on the rider-specific bicycle function algorithm. This can advantageously optimize the control of the bicycle and/or the bicycle component. The bicycle function algorithm can be stored locally, for example in a storage unit of the bicycle or in a storage unit of the bicycle component. It is also conceivable that the bicycle function algorithm is stored decentrally, for example in a cloud. The method for determining the bicycle function algorithm preferably takes place in a cloud or another type of computing network or on a server. The sensor data can be recorded directly from the bicycles and/or indirectly from bicycle components. The metadata is preferably information that cannot be recorded via sensors on the bicycle or sensors on the bicycle component. The metadata can, for example, be queried by the bicycle and/or the bicycle component and recorded via user input. It is also conceivable that the metadata is provided via a database. The driver types can be, for example, a sporty driver type, a safety-conscious driver type, a risk-taking driver type, a comfort-conscious driver type, a fitness-oriented driver type, a speed-oriented driver type, an acceleration-oriented driver type, etc. in preferably different gradations. The bicycle can be designed in particular as an electric bicycle. In the context of this application, an electric bicycle is to be understood in particular as a bicycle which has a drive unit to assist the rider. The electric bicycle is preferably designed as an e-bike, a pedelec, a cargo bike, a folding bicycle or the like. The drive unit has a motor which can be designed, for example, as a mid-engine or as a hub motor. The motor is preferably designed as an electric motor. The drive unit is connected to an energy storage device for supplying the drive unit with energy. The energy supply unit is preferably designed as a battery pack and has a battery housing which is preferably detachably connected to a frame of the bicycle. The electric bicycle comprises electronics with a control unit for controlling or regulating the electric bicycle. The electronics preferably comprise a sensor unit, wherein the sensor unit can have, for example, motion sensors, torque sensors, speed sensors, a GNSS receiver, magnetic sensors or the like. In addition, the electronics comprise a communication interface for wirelessly connecting the electric bicycle to an external device, such as a smartphone, and/or a server. The bicycle component can, for example, be used as a bicycle light, as a damping system, as an anti-lock braking system, as a seat post, as a gearshift or the like. It is also conceivable that the bicycle component is designed as an external device, in particular a mobile terminal, such as a smartphone, wherein the smartphone can preferably be connected to the bicycle during use. The determination of the rider type and the determination of the bicycle function algorithm can be based on a machine learning system. In the context of this application, a machine learning system is understood to mean in particular algorithms that build a statistical model using training data. The statistical model can be used to determine, for example, parameters and attributes that go beyond the scope of the training data. The algorithms of the machine learning system can be algorithms for supervised learning, unsupervised learning or reinforcement learning. The machine learning system can be designed as a neural network, for example. The training data, in particular the sensor data and the metadata, are preferably recorded by the bicycle and made available for training the machine learning system. However, it is also conceivable that the training data is partially or completely recorded and/or made available by an external device. The machine learning system is preferably trained on a server or the server. It is also conceivable that the machine learning system is also trained locally on the vehicle or the external device and that the large number of trained machine learning systems are then brought together in the server. The method according to th