DE112020007528T5 - Device and method for driving monitoring - Google Patents

Abstract

Method and device (100) for driving monitoring, wherein the device (100) means (108) configured to receive sensor data, means (110) configured to estimate a driving trajectory of a vehicle, and means (114) that configured to analyze the sensor data, wherein the means (108) configured to receive sensor data, the means (110) configured to estimate the travel trajectory of the vehicle, and the means (114) configured to analyze the sensor data and the driving path are configured to receive sensor data from at least one sensor of a vehicle, to determine a driving trajectory of the vehicle depending on the sensor data, to provide a reference for driving according to a vehicle type, a driving mode for the vehicle or a driving style, to determine an award depending on a difference between the driving trajectory and the reference, and to issue the award.

Description

The invention relates to a device and a method for driving monitoring.

US20180060970A1 discloses a driver assistance system for collision avoidance by analyzing the driving behavior in relation to critical traffic situations before which the driver can be warned. US20170166217A1 and WO 2014029882 A1 disclose aspects of customizing appropriate alerts according to different vehicle types, driving modes and styles.

The method and the device according to the independent claims further improve the driving monitoring.

The method for driving monitoring includes receiving sensor data from at least one sensor of a vehicle, determining a driving trajectory of the vehicle depending on the sensor data, providing a reference for driving according to a vehicle type, a driving mode for the vehicle or a driving style, determining a award depending on a difference between the traveling trajectory and the reference and issuing the award.

The method advantageously includes determining at least one predicted trajectory of a road user using the sensor data, with the reference being determined as a function of the at least one predicted trajectory.

In one aspect, the award is determined as a function of an artificial intelligence model, the artificial intelligence model being trained with different driving styles from recorded driving, particularly by professional drivers.

The artificial intelligence model can be trained to classify the driver's driving style, in particular as sporty, aggressive, safe, ecological, the method further comprising determining the reference to driving depending on the driving style.

The method advantageously includes outputting the award on a graphical user interface of the vehicle or on a social media interface.

The method advantageously includes a determination of a large number of differences between different travel trajectories and respective references for different driving situations and a determination of the award as a function of the large number of differences.

Advantageously, the method includes determining a goal depending on the vehicle type, driving mode for the vehicle, or driving style, determining whether the difference meets the goal, providing the award when the goal is met, and not providing the award otherwise .

The method may include determining the driving mode selected by a driver of the vehicle via a user interface.

The device for driving monitoring comprises means that are configured to receive sensor data, means that are configured to estimate a driving trajectory of a vehicle, and means that are configured to analyze the sensor data, the means that are configured to sensor data received, the means configured to estimate the driving trajectory of the vehicle and the means configured to analyze the sensor data and the driving trajectory are configured to cooperate to perform steps of the method.

• 1 schematisch eine Vorrichtung zur Fahrüberwachung,
• 2 Schritte in einem Verfahren zur Fahrüberwachung.

Further advantageous embodiments result from the following description and the drawing. Show it:

• 1 schematically a device for driving monitoring,
• 2 Steps in a driving monitoring method.

1 shows a device 100 for driving monitoring. In the example, the device 100 can be connected to at least one first sensor 102, at least one second sensor 104 and at least one third sensor 106 or comprises them. The at least one first sensor 102 is a camera in the example. The at least one second sensor 104 is a radar sensor in the example. The at least one third sensor 106 is a LIDAR sensor in the example. Other sensors can also be used. More or less than three sensors can be provided.

The device 100 includes a first module 108 configured to receive sensor data, a second module 110 configured to estimate a travel path of a vehicle, a third module 112 configured to monitor a journey, and a fourth module 114 configured to analyze the sensor data and the driving path.

The device 100 may be mountable on the vehicle. The device 100 may be a controller for the vehicle. The resources of the device 100 may be distributed across various controllers mounted on the vehicle and configured to communicate with each other.

These modules are configured according to a method described below for receiving sensor data from at least one sensor of a vehicle, determining a driving trajectory of the vehicle depending on the sensor data, providing a reference for driving according to a vehicle type, a driving mode for the vehicle or a driving style, determining an award depending on a difference between the traveling trajectory and the reference and issuing the award.

The first module 108 is configured to receive inputs from the at least one first sensor 102, e.g. B. the camera to receive. The first module 108 is further configured to receive inputs from the at least one second sensor 104, e.g. B. the radar sensor sensors to receive. The first module 108 is further configured to receive inputs from the at least one third sensor 106, e.g. B. the LIDAR sensor to receive.

The first module 108 can be configured to determine merged sensor data based on the sensor data from different sensors. The first module 108 may be configured to determine an object list that includes one or more objects detected in the sensor data received from the sensors. The first module 108 can be configured to determine separate object lists for different sensors.

The first module 108 can be configured to provide a first input to the second module 110 and to provide a second input to the third module 112 .

The first input can be fused sensor data. The second input may include the object list or the separate object lists and may include other features of the fused sensor data.

The second module 110 is configured to receive additional sensor data, e.g. B. steering wheel angle, acceleration or other vehicle parameters to receive.

The second module 110 is configured to calculate a future route based on the additional sensor data, e.g. B. the steering wheel angle, the acceleration or the other vehicle parameters to estimate.

The second module 110 may include a model for estimating the future travel path. The model may be an artificial intelligence based model. This artificial intelligence-based model can be trained to predict future route estimates based on additional sensor data.

The first input from the first module 108 can be an input, in particular into this model based on artificial intelligence, for estimating the future route.

The second module 110 may be configured to use map data to determine road data and to estimate the future travel path depending on the road data.

The second module 110, in particular the model, can be parameterized based on a vehicle type and/or a selected driving mode. The driving mode can be a normal mode, a sport mode or an ecological mode.

The future route, which is determined by the second module 110, is made available to the third module 112.

The third module 112 is configured to determine a predicted path for at least one road user. The third module 112 is configured to calculate the predicted path based on the second input, i. H. the object list or object lists from the sensors. In the example, a plurality of predicted paths for the driving path of the road users, i. H. of the object or objects detected in the sensor data.

In the example, the third module 112 is configured to determine a predicted trajectory with corresponding uncertainty for each detected object.

The third module 112 may be configured to determine a future trajectory of the vehicle based on the future travel path of the vehicle.

The third module 112 is configured to evaluate the predicted trajectory of at least one object and the future trajectory of the vehicle to determine a parameter indicative of a safety of driving. The evaluation can take into account a possible future collision if the future trajectory of the vehicle and the predicted trajectory of at least one object intersect. The evaluation can use the parameter ter so that in this case it indicates a high risk of collision. The evaluation may determine the parameter to indicate a risk of collision if a distance between these trajectories is less than a threshold without crossing them. A high acceleration or high speed of the at least one object or the vehicle can increase the risk compared to a lower acceleration or speed. The parameter can be determined to indicate a higher degree of risk of collision at a higher acceleration or speed than when the acceleration or speed is lower. A stability of the vehicle can be determined and the degree of risk can be set higher in an unstable driving situation of the vehicle than in a stable driving situation of the vehicle.

The parameter can also define the degree of risk depending on the vehicle type and/or the selected driving mode. The sport mode can cause the parameter to indicate a higher risk level or a lower risk level than the normal mode or the ecological mode.

In the example, the parameter can define three levels of risk, from high to low: Imminent, Critical, and Standard.

The third module 112 may be configured to signal a driver the criticality of the future estimated behavior, e.g. B. with the three degrees: Imminent, Critical and Standard.

The third module 112 can be configured to output data to the fourth module 114 . The data can be the predicted trajectory of at least one object and the future trajectory of the vehicle and/or the parameter indicative of a safety of driving.

The fourth module 114 is configured to analyze the predicted trajectory of at least one object and the future trajectory of the vehicle. The fourth module 114 is configured in the example to determine and display a safety analysis for a scenario in which the vehicle is driving. The fourth module 114 may be configured to display a history of trips for individual drivers of the vehicle. The history can be a scenario-based safety analysis or show a development of the safety analysis for the individual driver. The fourth module 114 can be configured to determine the security analysis using an artificial intelligence. The artificial intelligence can be trained to predict target achievements. The fourth module 114 may also be configured to determine data analysis for vehicle malfunction possibilities.

The fourth module 114 may be configured to analyze the predicted trajectory of at least one object and the future trajectory of the vehicle and the at least one parameter indicative of safety of driving.

The fourth module 114 is configured in one aspect to provide a safe driving assessment, e.g. B. for a single driver to generate. The fourth module 114 may be configured to upload the assessment to a cloud-based unsafe behavior learning and assessment tool.

In another aspect, the fourth module 114 is configured to use an external application or applications to simulate various behaviors. In another aspect, the fourth module 114 is configured to use an external application or applications to present safe and unsafe situations.

The fourth module 114 may be configured to implement a gamification concept for awards. For example, an award is given for a driver driving the vehicle 100 km without warnings. For example, an award is given for a flawless overtaking.
The award may include a badge displayed to the driver via a vehicle graphical user interface.

The artificial intelligence can be used for the prediction and/or the analysis of the data as follows.

For prediction:

An artificial intelligence model can be trained with general driver behavior to accurately predict how other road users will behave.

A brief history of a vehicle's motion along with current dynamic attributes, such as speed, acceleration, yaw, yaw rate, can be input into the artificial intelligence model to predict the vehicle's future trajectory. A sequential model can be used, e.g. B. an LSTM (Long Short-Term-Memory).

For analysis:

An artificial intelligence model can be trained with different driving styles of recorded professional drivers.

The model can be used to classify the driver's driving style, e.g. B. sporty, aggressive, safe, ecological, are used.

The model can output a "driving style" difference to assess how close a driver is to a desired driving style. The model can issue a "rating".

The fourth module 114 may be configured to use this "score" to determine award or suggest improvements to driving.

An improvement is, for example, a recommendation for a driver of the vehicle to increase a distance to another vehicle driving ahead, to brake earlier in tight corners.

The model can be trained using famous drivers. The model can issue a suggestion or recommendation for achieving a driving style that most closely resembles the famous driver. The model may issue the award depending on the difference from the famous driver's driving style, e.g. B. issue a larger award, the better the driving style is imitated. The output can be made available for sharing on social media.

Depending on the vehicle type, selected driving mode or driving style, a different award can be determined.

The method for driving monitoring is described below with reference to 2 described.

In a step 202, sensor data from at least one sensor of the vehicle are provided. The sensor data may be collected while the vehicle is moving in a drive cycle.

In a step 204 the travel path of the vehicle is determined. In the example, the travel trajectory is determined using the sensor data, as described above. The at least one predicted trajectory of the road user can also be determined on the basis of the sensor data.

In a step 206, the reference for driving is provided according to the vehicle type, the driving mode for the vehicle, or the driving style, as described above. The reference can also be determined as a function of the at least one predicted trajectory. The reference can be determined by the artificial intelligence model.

The artificial intelligence model can be trained to classify the driver's driving style, in particular as sporty, aggressive, safe, ecological. In this case, the reference for driving can be determined depending on the driving style into which the artificial intelligence model has classified the sensor data.

In a step 208, the award is determined as a function of the difference between the driving trajectory and the reference, as described above. In the example, the award is determined depending on the artificial intelligence model.

The artificial intelligence model can be trained with different driving styles, especially recorded professional drivers.

A large number of differences between different travel trajectories and respective references can be determined for different driving situations. This can happen within the driving cycle or for the same driver in different driving cycles for the vehicle. In this case, the award is determined depending on the large number of differences, e.g. B. by summing up individual awards intended for the various driving situations.

A destination may be provided depending on the vehicle type, the driving mode for the vehicle, or the driving style. In this case, the award can be provided if the difference satisfies the goal. In one aspect, if the goal is not met, the award is not provided.

In a step 210, the award z. B. output on a graphical user interface of the vehicle or to a social media interface. The driving score can also be determined and uploaded to the cloud-based tool.

The driving mode may be selectable by a driver in the vehicle via a user interface. In this case, the driving mode selected by a driver of the vehicle can be recognized for use in the method.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• US20180060970A1[0002]
• US20170166217A1[0002]
• WO 2014029882 A1 [0002]

Claims (9)

Method for driving monitoring, characterized by receiving (202) sensor data from at least one sensor of a vehicle, determining (204) a driving trajectory of the vehicle depending on the sensor data, providing (206) a reference for driving according to a vehicle type, a driving mode for the vehicle or a driving style, determining (208) an award depending on a difference between the driving trajectory and the reference, and issuing (210) the award. procedure after claim 1 , characterized by determining (204) at least one predicted trajectory of a road user based on the sensor data, the reference being determined as a function of the at least one predicted trajectory (206). Method according to one of the preceding claims, characterized in that the award is determined as a function of an artificial intelligence model (208), the artificial intelligence model being trained with different driving styles from recorded journeys, in particular by professional drivers. procedure after claim 3 , characterized in that the artificial intelligence model is trained to classify the driver's driving style, in particular as sporty, aggressive, safe, ecological, the method further comprising determining the reference to driving depending on the driving style. Method according to one of the preceding claims, characterized by outputting (210) the award on a graphical user interface of the vehicle or on a social media interface. Method according to one of the preceding claims, characterized by determining (208) a large number of differences between different travel trajectories and respective references for different driving situations and determining (208) the award as a function of the large number of differences. Method according to one of the preceding claims, characterized by determining a target depending on the vehicle type, the driving mode for the vehicle or the driving style, determining whether the difference meets the target, providing the award if the target is met, and not providing the award otherwise. Method according to any one of the preceding claims, characterized by determining the driving mode selected by a driver of the vehicle via a user interface. Device (100) for driving monitoring, characterized in that the device (100) has means (108) configured to receive sensor data, means (110) configured to estimate a driving trajectory of a vehicle, and means (114) configured to analyze the sensor data, the means (108) configured to receive sensor data, the means (110) configured to estimate the travel path of the vehicle, and the means (114) , configured to analyze the sensor data and the driving trajectory, configured to cooperate to perform steps of the method according to any one of the preceding claims.

Images