GB2599710A

GB2599710A - A method for determining a personalized estimated time of arrival of a motor vehicle as well as a corresponding assistance system

Info

Publication number: GB2599710A
Application number: GB2016064.4A
Authority: GB
Inventors: Hsu Ting-Wei
Original assignee: Daimler AG
Current assignee: Mercedes Benz Group AG
Priority date: 2020-10-09
Filing date: 2020-10-09
Publication date: 2022-04-13
Also published as: GB202016064D0

Abstract

A method for determining a personalized estimated time of arrival (pETA) for a route 18 of a motor vehicle 10. An error value is determined by comparing the current estimated time of arrival (predicted on different contexts of an individual user behaviour) with the average estimated time of arrival (aETA). The error value is then used to train a personalised model 26 and the personalized estimated time of arrival (pETA) is determined by the electronic computing device (16) by using the trained personalized model (26). The electronic computing device may comprise a neural network which is used to train the personalized model. The characterizing parameter of the surroundings of the motor vehicle may be taken into consideration for training the personalized mode. The average estimated time of arrival may be estimated by summing an estimated time of the road segments and intersections making up the route. Furthermore, the invention relates to a corresponding assistance system (12).

Description

A METHOD FOR DETERMINING A PERSONALIZED ESTIMATED TIME OF ARRIVAL OF A MOTOR VEHICLE AS WELL AS A CORRESPONDING ASSISTANCE SYSTEM

FIELD OF THE INVENTION

[0001] The invention relates to the field of automobiles. More specifically, the invention relates to a method for determining a personalized estimated time of arrival for a route of a motor vehicle by an assistance system of the motor vehicle, as well as a corresponding assistance system.

BACKGROUND INFORMATION

[0002] Presently, most transportations, for example, car, boat, or aircraft, has an average estimated time of arrival (ETA) within their navigation systems. Although the average ETA sometimes may be enhanced by environmental factors, such as road conditions, weather, or furthermore, the ETA is not personalized for individual users. Furthermore, a dynamic model of a driver control strategy of lane change behavior and trajectory planning for collision prediction is already known from the state of the art.

[0003] The scientific publication "Dynamic Modeling of Driver Control Strategy of Lane-Change Behavior and Trajectory Planning for Collision Prediction" from Guoqing Xu et. al., published in IEEE Transactions on Intelligent Transportation Systems, Volume: 13, Issue: 3, Sept. 2012, introduces a dynamic model of the driver control strategy of lane-change behavior and applies it to trajectory planning in driver-assistance systems. The proposed model reflects the driver control strategies of adjusting longitudinal and latitudinal acceleration during the lane-change process and can represent different driving styles by using different model parameters.

SUMMARY OF THE INVENTION

[0004] It is an object of the invention to provide a method as well as a corresponding assistance system, by which a personalized time of arrival may be computed in a motor vehicle.

[0005] This object is solved by a method as well as a corresponding assistance system according to the independent claim. Advantageous embodiments are presented in the subclaims.

[0006] One aspect of the invention relates to a method for determining a personalized estimated time of arrival for a route of a motor vehicle by an assistance system of the motor vehicle, wherein an average estimated time of arrival for the route is provided by the assistance system, wherein a current user behavior of a user of the motor vehicle is detected by a detection device of the assistance system, and wherein the personalized estimated time of arrival is determined depending on the average estimated time of arrival and the detected user behavior by an electronic computing device of the assistance system.

[0007] In one embodiment based on the user's behavior, a current estimated time of arrival for the route is stored in a storing device of the electronic computing device and the current estimated time of arrival is compared with the average estimated time of arrival and an error value is determined depending on the comparison, wherein a personalized model for the personalized estimated time of arrival is trained depending on the error value. Further, the personalized estimated time of arrival may also be determined by the electronic computing device by using the trained model.

[0008] Therefore, a more accurate personalized estimated time of arrival for the user is presented, wherein this may save time, money and improve the user experience and help the user with decision making. For example, the user who drives faster on a road may choose the route with less traffic lights but with a slightly longer route to reach the destination more quickly.

[0009] Therefore, the personalized estimated time of arrival is computed by taking the individual behavior of the user under different environmental conditions into consideration.

For example, the user may drive more slowly or the same when it is raining. Therefore, a model is created which learns individual user behavior when the user is using the transportation. The model starts with the average estimated time of arrival and gradually adjusts it to the user's behavior over time with the integration of decision theories and features in machine learning systems for user predictions.

[0010] According to an embodiment, the electronic computing device comprises a neural network and the personalized model is trained by using the neural network.

[0011] In another embodiment, a characterizing parameter of surroundings of the motor vehicle is detected by a further detection device of the assistance system and the characterizing parameter of the surroundings are taken into consideration for training the personalized model by the electronic computing device.

[0012] In another embodiment for providing the average estimated time of arrival at least one route is subdivided into at least one road segment and at least one intersection, and an estimated time of the at least one road segment and of the at least one intersection summed up to a total time value for the route, and the average estimated time of arrival is determined depending on the total time value.

[0013] Another aspect of the invention relates to an assistance system for determining a personalized estimated time of arrival of a motor vehicle, with at least one detection device and one electronic computing device, wherein the assistance system is configured to perform a method according to the preceding aspect. In particular, the method is performed by the assistance system.

[0014] A still further aspect of the invention relates to a motor vehicle with an assistance system according to the preceding aspect. The motor vehicle may be configured as a car, a boat, or an aircraft.

[0015] Advantageous forms of configuration of the method are to be regarded as advantageous forms of the assistance system as well as the motor vehicle. The assistance system as well as the motor vehicle therefore comprise means for performing the method.

[0016] Further advantages, features, and details of the invention derive from the following description of preferred embodiments as well as from the drawings. The features and feature combinations previously mentioned in the description as well as the features and feature combinations mentioned in the following description of the figures and/or shown in the figures alone can be employed not only in the respectively indicated combination but also in any other combination or taken alone without leaving the scope of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The novel features and characteristics of the disclosure are set forth in the independent claims. The accompanying drawings, which are incorporated in and constitute part of this disclosure, illustrate exemplary embodiments and together with the description, serve to explain the disclosed principles. In the figures, the same reference signs are used throughout the figures to refer to identical features and components. Some embodiments of the system and/or methods in accordance with embodiments of the present subject matter are now described below, by way of example only, and with reference to the accompanying figures.

[0018] Fig. 1 shows in a schematic view a graph for a route according to an embodiment of an assistance system: and [0019] Fig. 2 shows another schematic view of a graph according to an embodiment of the assistance system.

[0020] In the figures the same elements or elements having the same function are indicated by the same reference signs.

DETAILED DESCRIPTION

[0021] In the present document, the word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0022] While the disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and will be described in detail below. It should be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the scope of the disclosure.

[0023] The terms "comprises'', "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion so that a setup, device or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a system or apparatus preceded by "comprises" or "comprise" does not or do not, without more constraints, preclude the existence of other elements or additional elements in the system or method.

[0024] In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form part hereof, and in which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.

[0025] Fig. 1 shows a schematic view of a motor vehicle 10. The motor vehicle 10 may be configured as a car, an aircraft, or a boat. The motor vehicle 10 comprises an assistance system 12. The assistance system 12 comprises at least one detection device 14 and one electronic computing device 16.

[0026] In Fig. 1, a route 18 is shown, wherein the route 18 comprises at least one road segment 20 as well as at least one intersection 22. According to the shown embodiment, the route 18 comprises five road segments 20 and six intersections 22.

[0027] According to Fig. 1, a method is presented for determining a personalized estimated time of arrival (pETA) of the motor vehicle 10 by the assistance system 12 of the motor vehicle 10, wherein an average estimated time of arrival (aETA) is provided by the assistance system 12, wherein a current user behavior of a user of the motor vehicle 10 is detected by the detection device 14 of the assistance system 12, and wherein the personalized estimated time of arrival pETA is determined depending on the average estimated time of arrival aETA and the detected user behavior by the electronic computing device 16 of the assistance system 12.

[0028] In an embodiment, based on the user's behavior, a current estimated time of arrival is stored in a storing device 24 of the electronic computing device 16 and the current estimated time of arrival is compared with the average estimated time of arrival aETA, and an error value is determined depending on the comparison, wherein a personalized model 26 for the personalized estimated time of arrival pETA is trained depending on the error value, and the personalized estimated time of arrival pETA is determined by the electronic computing device 16 by using the trained personalized model 26.

[0029] In an embodiment the electronic computing device 16 comprises a neural network 28, and the personalized model 26 is trained by using the neural network 28.

[0030] According to an embodiment of the assistance system 12, the personalized estimated time of arrival pETA is provided by traffic data by the states, service provider, phone users, and so on. The error value is predicted on different contexts of an individual user behavior from the average estimated time of arrival aETA. According to the embodiment shown in Fig. 1, there exist graphs 30 for the road segments 20 and the intersections 22. The graphs 30 are generated based on contexts and average speed for the road segments 20 and based on density and time for intersections 22. For each road segment 20, there exists a collected graph distribution of how contexts affect the speed. The contexts may be a time of day, a day of the week, events nearby, or holidays. The context may be multiple dimensions. For example, at 9 a.m. on Monday, it is known that this road segment 20 of the road will be busy for instance due to the usually heavy traffic at this time of the weekday. There are data to keep track of that road segment 20 and how long it takes the traffic participants to pass the road segment 20. This is a function of context. For each intersection 22, there exists a collected graph distribution of how density affects time spent at the intersection 22. For example, data may indicate that traffic participants drive up to the intersection 22 when it is green and drive through immediately. Another example is that traffic participants drive up to the intersection 22 when it is red and have to wait for some time. A further example is that people may realize that there is a line at the intersection 22 and wait there for an extended period of time.

[0031] The graph data for road segments 20 and intersections 22 are available throughout the whole map collected by traffic data by the states, service providers, and phone users.

[0032] The average estimated time of arrival aETA may be computed for any route indicated in Fig. 1. For those road segments 20, a distribution of the route 18 may be computed, the current contexts may be matched, and the average estimated time of arrival aETA may be provided. For the intersection 22, the distribution of the intersection 22 and the average estimated time to wait at the intersection 22 may be computed by matching the current contexts. These estimated times may be summed up in order to obtain the average estimated time of arrival aETA. For example, according to Fig. 1, the average estimated time of arrival aETA may be 31 minutes.

[0033] Fig. 2 shows another schematic view of the route 18 according to Fig. 1. Fig. 2 shows that there is a possibility that there will be new road segments 32. The new road segments 32 may have very little data from service providers, wherein a kinematic solution of time travel on the road segment 20 estimating the time may be computed. The general approach for personalizing the estimated time of arrival by public is to collect more data on the individual users and build the graph 34 of the road segment 20 and/or the intersection 22 of that specific user. The assistance system 12 is configured to build personalized data not only on a specific route but for the entire map including places the user has never visited before, using the average estimated time of arrival aETA as a base. It is provided that the electronic computing device 14 predicts how long the average estimated time of arrival aETA is away from the current estimated time of arrival. Therefore, the error is computed between the average estimated time of arrival aETA and the current estimated time of arrival in particular for multiple routes. The error further be used in the neural network 28 to have a more accurate pETA of the specific user.

[0034] For example, if the user typically arrives five minutes earlier than the average estimated time of arrival aETA computes, this is learned by the neural network 28. Therefore, the error value can be X%, that the user is X% faster or, for example, slower than the average estimated time of arrival aETA. As an example, the route 18 has an average estimated time of arrival aETA of 31 minutes and the current estimated time of arrival is 29 minutes. Then the personalized model 26 learns that the user needs 93.5% of the average estimated time of arrival aETA. A user may drive faster in general, because he may be more aggressive and faster than the speed limit. The error value in percentage is computed and on all collected road segments 20 and intersections 22 of this user. Here, for example, the percentage of the end result of the behavior is computed instead of looking at the specific behavior, for example, a lane change style, speed, a brake pressing. Therefore, it is not learned what kind of behavior results from faster or slower driving versus the neural network 28 learns how much faster or slower the user is than the average.

[0035] In particular, a characterizing parameter of the surroundings of the motor vehicle 10 is detected by a further detection device of the assistance system 12 and the characterizing parameter of the surroundings are taken into consideration for training the personalized model 26 by the electronic computing device 16. For example, during rush hour, if the user is stuck in a traffic jam and cannot drive as fast as he normally would. So he is 0% faster or slower than the average. During weekends, however, the same user may be 4% faster than the average estimated time of arrival aETA. The personalized model 26 learns based on the context for the specific user and obtains the error value for the average estimated time of arrival aETA. Therefore, the percentage prediction may be multiple dimensions.

[0036] Once, the error value is learned, the electronic computing device 16 applies this to the entire map based on the average estimated time of arrival aETA. Also, in the case of cold start problems, the error value is 1.

Personalized estimation = Average ETA *% Error of diff context [0037] Fig. 1 and Fig. 2 show a user action prediction with behavioral theories.

Reference Signs motor vehicle 12 assistance system 14 detection device 16 electronic computing device 18 route road segment 22 intersection 24 storing device 26 personalized model 28 neural network graph 32 new road segment aETA average estimated time of arrival pETA personalized estimated time of arrival

Claims

CLAIMS1. A method for determining a personalized estimated time of arrival (pETA) for a route (18) of a motor vehicle (10) by an assistance system (12) of the motor vehicle (10), wherein an average estimated time of arrival (aETA) for the route (18) is provided by the assistance system (12), wherein a current user behavior of a user of the motor vehicle (10) is detected by a detection device (14) of the assistance system (12), and wherein the personalized estimated time of arrival (pETA) for the route (18) is determined depending on the average estimated time of arrival (aETA) and the detected user behavior by an electronic computing device (16) of the assistance system (12), characterized in that based on the user behavior a current estimated time of arrival for the route (18) is stored in a storing device (24) of the electronic computing device (16) and the current estimated time of arrival is compared with the average estimated time of arrival (aETA) and an error value is determined depending on the comparison, wherein a personalized model (26) for the personalized estimated time of arrival (pETA) is trained depending on the error value and the personalized estimated time of arrival (pETA) is determined by the electronic computing device (16) by using the trained personalized model (26).
2. The method according to claim 1, characterized in that the electronic computing device (16) comprises a neural network (28) and the personalized model (26) is trained by using the neural network (28).
3. The method according to claim 1 or 2, characterized in that a characterizing parameter of surroundings of the motor vehicle (10) is detected by a further detection device of the assistance system (12) and the characterizing parameter of the surroundings are taken into consideration for training the personalized model (26) by the electronic computing device (16).
4. The method according to any one of claims 1 to 3, characterized in that for providing the average estimated time of arrival (aETA) at least one route (18) is subdivided into at least one road segment (20) and at least one intersection (22), and an estimated time of the at least one road segment (20) and of the at least one intersection (22) are summed up to a total time value for the route (18) and the average estimated time of arrival (aETA) is determined depending on the total time value.
5. An assistance system (12) for determining a personalized estimated time of arrival (pETA) of a motor vehicle (10), with at least one detection device (14) and one electronic computing device (16), wherein the assistance system (12) is configured to perform a method according to anyone of claims 1 to 4.