GB2594491A

GB2594491A - Passenger grouping prediction for autonomous ride sharing for an optimal social experience

Info

Publication number: GB2594491A
Application number: GB2006336.8A
Authority: GB
Inventors: Voigt Martin; Hill Andrew
Original assignee: Daimler AG
Current assignee: Mercedes Benz Group AG
Priority date: 2020-04-30
Filing date: 2020-04-30
Publication date: 2021-11-03
Also published as: GB202006336D0

Abstract

A method and system for forming ride-sharing groups of one or more riders is provided. A ride sharing request from each rider is received, wherein the ride sharing request comprises source and destination of the requested ride, and one or more demographic details of each rider. Social attitude parameters of each rider are also received. A context information of each rider is determined, and a score of each rider based on the received ride sharing request and the context information is calculated. Based on the comparison of the calculated score of each rider, the processor is configured to predict ride-sharing groups comprising at least two riders. The social attitude parameters may be determined from social networking profiles, sensors inside or outside a rider’s vehicle, questionnaires and/or shared post ride experiences.

Description

THE PATENTS ACT, 1970 (39 of 1970) The patent Rule, 2003

COMPLETE SPECIFICATION

(See section 10 and rule 13)

TITLE OF THE INVENTION

PASSENGER GROUPING PREDICTION FOR AUTONOMOUS RIDE SHARING FOR AN OPTIMAL SOCIAL EXPERIENCE

Name and address of the applicant: a) Name: Daimler AG b) Nationality: Get-many c) Address: 70372, Stuttgart, Get-many.

[0001] PREAMBLE TO THE DESCRIPTION

[0002] The following specification particularly describes the invention and the manner in which it is to be performed:

[0003] TECHNICAL FIELD

[0004]The present subject matter relates to the field of automobiles. Particularly, the present subject matter relates to grouping of riders in a ride sharing system.

[0005] BACKGROUND

[0006] People generally like to share a ride with someone they know so that they can enjoy social interaction. While using a public transport, people usually share the ride with unknown travelers and there is a chance that a rider may have an unpleasant experience due to undesirable travelers. A rider may have a nature of socializing with others while the other rider may like to remain silent and not like to interact with fellow riders. If these two riders are matched together in a ride sharing system, they may not experience a good ride. In another example, a rider might be engaged in a task such as reading a book, working on his computer, etc. and thus might not like to be disturbed by anyone. If such a rider is bothered by another rider who insists on interaction, then such a ride may give an unpleasant experience.

[0007] Thus, there is a need in the art for grouping riders together such that the riders get a pleasant experience while sharing rides.

[0008] SUMMARY

[0009] In one non-limiting example, a method and a system forming ride-sharing groups of one or more riders is provided. Firstly, a ride sharing request from each rider is received by a transceiver, wherein the ride sharing request comprises source and destination of the requested ride, one or more demographic details of each rider.

The transceiver further receives social attitude parameters of each rider. A context information of each rider is determined by a processor. The processor calculates a score of each rider based on the received ride sharing request and the context 10 15 20 information. Based on the comparison of calculated score of each rider, the processor is configured to predict ride-sharing groups which comprises at least two riders. This ensures that the group customization of riders in the ride sharing system causes a pleasant riding experience.

[00101 The present disclosure overcomes one or more shortcomings of the prior art and provides additional advantages discussed throughout the present disclosure. Additional features and advantages are realized through the techniques of the present disclosure. Other embodiments and aspects of the disclosure are described in detail herein and are considered a part of the claimed disclosure.

[00111 BRIEF DESCRIPTION OF THE DRAWINGS

[00121 The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed embodiments. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the figures to reference like features and components. Some embodiments of system and/or methods in accordance with embodiments of the present subject matter are now described, by way of example only, and with reference to the accompanying figures, in which: [00131 Figure 1 depicts an exemplary block diagram of a system in accordance with one embodiment of the present invention.

[00141 Figure 2 depicts an exemplary block diagram of a system in accordance with one embodiment of the present invention.

100151 Figure 3 depicts an exemplary embodiment in accordance with present invention.

[00161 Figure 4 depicts an exemplary embodiment in accordance with present invention.

[00171 Figure 5 depicts method in accordance with one embodiment of the present invention.

[00181 It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will he appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes which may he substantially represented in computer readable medium and 0 executed by a computer or processor, whether or not such computer or processor is explicitly shown.

[00191 DETAILED DESCRIPTION

[00201I11 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.

[00211 In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form a 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.

[0022] The present disclosure addresses the shortcomings of the conventional art and proposes a method and an apparatus for predicting grouping of riders in a ride sharing system such that the riders get a pleasant riding experience.

[0023] Referring to figure 1, an exemplary block diagram of a system 100 for forming ride-sharing groups of one or more riders is provided. The system 100 comprises a memory 104 for storing instructions. The system further comprises a transceiver 106 configured to transmit/receive data. The system further comprises a processor 102 coupled to the memory 104 and the transceiver 106. The processor 102 is also coupled to one or more sensors placed inside and outside a ride sharing vehicle.

[0024] Each rider 120 may have a user equipment 122 which may be used by the one or more riders 120 to book a ride. The user equipment 122 may be any computing device, for example, but not limited to a mobile phone, tablet, computer, laptop. The user equipment 122 may communicate with the system 100 through a network 112. The user equipment 122 may also have a display to display an application through which a rider sends a ride-sharing request to the system 100. The application may include demographic details of the rider 120. The demographic details may be pre-stored by the rider 120 when an account is created in the application by the rider 120.

The demographic details may include, for example, but not limited to name, age, sex, address, nationality of the rider 120. In one embodiment, the demographic details may be stored in a memory of the user equipment. In another embodiment, the demographic details may be stored in the memory 104 of die system 100.

[0025] The system 100 may have access to the current location of the rider. The rider 120 may input a place of origin from which the rider 120 wants to take the ride. The terms source and origin have been used interchangeably in rest of the specification. In another embodiment, the rider may want to take a ride from the current position. In such a case, the system 100 may automatically detect the current location of the rider as source location. The rider may also input a place of destination where the rider 120 would like to be dropped.

[0026] The transceiver 106 receives a ride sharing request from each rider 120. The ride sharing request includes one or more demographic details, the information about the origin and the destination of each rider. The total number of riders to be matched depends on the total capacity available inside the ride sharing vehicle. For example, if the ride sharing vehicle is a 4-seater vehicle, the total number of riders to be matched together may be 2, 3 or 4.

[0027] In addition to the ride sharing request, the system 100 uses context information to predict grouping of the riders together. The context information is as follows: the processor 102 is configured to determine a context information of each rider 120. To determine the context information, the transceiver 106 is configured to receive social attitude parameters of each rider 120. The social attitude parameters may express how socially active or inactive a rider is. The term socially active denotes if a rider 120 likes to socialize with other riders/people. This may include engaging in talks with other people, visiting public places for example, concerts, clubs, etc. The social attitude parameters may also include how much a rider is active on his social media profiles. The social attitude may also be set up manually by the rider 120 himself ahead of the time or at the time of the ride sharing experience.

[0028] In one embodiment, a rider 120 may be asked several questions related to his personal interests via the user equipment 122. The questions may include the kind of person the rider may like to share his ride, the kind of songs the rider likes to listen, the hobbies or interests of the rider, etc. These questions may be asked at a time the rider subscribes to the ride sharing service. In another embodiment, the questions may be asked at any future time. As soon as the rider 120 replies to the questions, the replies may be stored in the memory of the user equipment 122 of the rider 120. In another embodiment, the inputs may be stored in the memory 104 of the system 100.

[0029] Another source of determining social attitude parameters is by obtaining post-ride experience of the rider 120. The post-ride experience may be obtained in the form of questions from the rider 120 and the rider 120 may be asked to input replies to the questions. The post-ride experience may include, for example, feedback/reviews related to driver or other riders, if the rider liked the conversation with the other rider, if he would like to ride again with the other rider, vehicle quality, etc. t00301 Another source of determining social attitude parameters may be a social network profile of the rider 120. The social network profile may include one or more social media accounts such as Facebook, histogram, Twitter, YouTube, Nettlix etc. of the rider 120. Through social network profile, the system may get access to social activity of the rider which may include, for example, but not limited to determining if the rider has liked any photo, commented on any photo, etc. In an exemplary scenario, the rider may have liked a photo or post relating to topic of politics. In such scenario, the processor 102 may determine that the rider 120 is interested in politics. In another exemplary scenario, the rider 120 may have commented on a topic related to cricket.

In such a scenario, the processor 102 may determine that the rider 120 is interested in cricket. The processor may base its result on a count of likes or posts by the rider related to a common topic. In another embodiment, die processor 102 may determine a frequency of activity of the rider 120 on the social media account by the rider 120. The frequency may include the number of times the rider is active on the social media account. With this, the processor 102 may determine how much the rider likes to socialize with other people/rider.

[00311 Another source of determining social attitude parameters is by way of input from one or more sensors present inside the vehicle. The one or more sensors may be, but not limited to, image capturing sensor, audio capturing sensor, emotion detecting sensor, eye movement sensor, smile detection sensor, global positioning sensor, heart monitor, blood pressure sensor, pulse sensor, motion sensor. The one or more sensors are in communication with the processor 102. The one or more sensors may detect the social attitude parameters of the rider 120 during the ride. The social attitude parameters may indicate how socially active the rider is during die ride.

[00321 In one exemplary embodiment, if the image capturing sensor determines that the rider 120 is engaged in conversation with another rider, the one or more sensors may communicate to the system 100, based on which the social attitude parameters may be determined. The processor 102 may categorize such a rider as socially active.

In another exemplary situation, if the image capturing sensor determines that the rider 120 is looking outside a window during the ride, the one or more sensors may communicate to the processor 102, based on which the social attitude parameters of the rider 120 may be determined. In another embodiment, the plurality of sensors may capture the scenes that the rider 120 is looking at outside the vehicle. These captured data may be used to start a conversation with the rider 120 or driver during the ride sharing experience.

[00331 The processor 102 may categorize a rider as socially inactive or active. Such categorization may be stored in the memory 104 for future reference. Based on the input received from the sensors inside the ride-sharing vehicle, the processor 102 may determine the amount of time the rider was conversing with the other rider and the amount of time the rider 120 was looking outside. The processor 102 may also determine the emotional state of the rider while conversing with the other rider through eye movement, the number of smiles the rider gave while conversing with the other rider, the pitch of the voice the rider had while conversing, the movements of the neck or the head while conversing, etc. Based on these movements, the processor 102 may determine the number of agreements or disagreements the riders had or the overall experience the riders had while sharing the ride.

[00341 Another source of determination of social attitude parameters of the rider 120 may be one or more sensors placed outside the vehicle or other data external to the vehicle. These one or more sensors can include smart home sensors that can communicate information to the vehicle, such as the time, frequency and nature of usage of home devices. For example a smart furnace or HVAC system, refrigerator, lighting system, etc. can transmit usage information to the vehicle. Additionally, smart cameras and/or speakers can communicate information about the number of people present in a home back to the vehicle, and this information can be sent individually or paired with e.g. smart doorbell system information to provide a potential determination on the likely number of passengers. Further, passenger social media account information such as incoming and outgoing messages, calendar appointments, postings and general activity can be communicated to the vehicle all to provide a more comprehensive context of ride sharing potential.

[0035] Once the social attitude parameters of each rider 120 are received from the above one or more sources, the processor 102 is configured to determine the context information of each rider 120. The context information may include a determination of how much socially active or inactive the rider is. This may determine the social attitude of the rider 120.

[0036] In order to determine the social attitude of the rider 120, the processor 102 is configured to calculate a score based on the context information and received ride sharing request. To calculate the score, the processor 102 is configured to express the context information in form of a continuous target variable. The continuous target variable may be a score assigned to each social attitude parameter. The score may be assigned on a scale of 0 and I where 0 represents that the rider 120 is not interested in socializing with anyone and 1 represents a strong desire for socialization.

[0037] The score may be 0 or 1 or anywhere between 0 and 1 depending on the social attitude parameters. For example, a score may be 0.6 if the rider 120 is happy to talk but doesn't seek it out or a 0.4 if the rider 120 likes to rather play with the phone but won't get angry if someone else is interacting with him, etc. Since it can be any real value between 0 and 1, it's continuous. The above examples are only for exemplary purpose and there may be numerous scenarios for assigning the score to the rider. The score of a rider may keep changing with new social attitude parameters being received by the system 100 in the manner explained above.

[0038] In another embodiment, a regression model may be prepared for each rider 120. The received social attitude parameters are fed in the regression model. The social attitude parameters are those defined above i.e. reply to questions received from the rider, post-ride experience received towards a ride by the rider, information extracted from social network profile(s) of the rider 120, information received from one or more sensors placed inside the ride sharing vehicle, and information received from one or more sensors placed outside the ride sharing vehicle. Once the social attitude parameters are fed inside the regression model, the regression model uses machine learning algorithms to calculate the score. Different regression techniques or loss functions are employed based on the fundamental quantities of the received social attitude parameters that are being modeled. In a non-limiting embodiment, for example, one could use a collaborative filtering approach. In such an example, a first user may give a similar post-ride summary to that of a second user, which may mean that these two users are similar. Therefore, actual scores computed in the past for the first user might be useful for predicting scores for the second user in the future. Such a similarity metric may be computed using singular value decomposition (SVD) on the sparse user pairing matrix, or more complex latent factor models can be used such as a variational autoencoder (VAE) using all demographic and context information combined with past post-ride summaries to predict future post-ride summaries while also building a user similarity metric in its architectural bottleneck. In this way, previous rides from multiple users with positive post-ride sentiment may imply that similarly responsive users would also experience a positive ride with each other, and a precise method of determining this similarity is provided by the machine learning algorithm. In one embodiment, the scores may be stored in the memory 104 and may be retrieved by the processor 102 to group the riders.

[00391 The regression model may be trained based on the one or more social attitude parameters. In one embodiment, the regression model may be updated regularly based on the social attitude parameters. For example, if during a plurality of rides, the rider is determined not to be socially active during the rides, the regression model is trained that the rider 120 is socially inactive.

[00401 Once the score is determined, the processor 102 is configured to predict grouping of the riders 120 together based on the social attitude parameters and ride sharing request.

10041111e score of each rider is compared with the score of another rider. For example, if the score of the rider is 0.6, the processor 102 may be configured to match the rider 120 with another rider having a score between 0.6 to 1. This is because the rider with a score of 0.6 is determined to be socially active, and another rider having a score of 0.6 or more would also be socially active. Overall, such grouping may help in riders having a pleasant ride experience.

[0042] Referring to figure 2, another embodiment in accordance with the present invention is shown. As shown in figure 2, the system 100 may receive (through transceiver 106) the ride sharing request and the social attitude parameters (in the manner as explained above) and send the ride sharing request and the social attitude parameters to a matching server 110. The matching server 100 may be a server located outside the system 100 and may be responsible for predicting grouping riders together in a ride based on the information received from the system 100.

[0043] Referring to figure 3, one exemplary embodiment is shown where the two riders are matched based on the ride sharing request and social attitude parameters. The figure shows two socially active riders are matched together and are engaged in a conversation in the ride sharing vehicle. In figure 3, both the riders matched inside the ride-sharing vehicle may have a common point of interest related to politics. Hence, it would be convenient for the riders to talk about the common interest during the ride.

[0044] Referring to figure 4, another exemplary embodiment is shown where two riders are matched based on the ride sharing request and social attitude parameters. The figure shows the riders are looking out of the vehicle as it was determined that both are socially inactive.

[0045] Referring to figure 5 a method 500 for grouping of riders is shown. At step 501, receiving a ride sharing request from each rider, wherein the ride sharing request comprises one or more demographic details of each rider, origin and destination of the requested ride of each rider. At step 503, the method comprises receiving social attitude parameters of each rider. At step 505, the method comprises determining a context information of each rider. At step 507, the method comprises calculating a score of each rider based on the received ride sharing request and the context information. At step 509, the method comprises predicting ride-sharing groups including at least two riders based on the calculated score.

[00461 The term 'rider" used throughout the complete specification would also include "driver" of the ride sharing vehicle and the techniques described herein to match different riders together would also be applicable on matching driver with the riders when one or more riders request a ride.

[00471 In one embodiment, the present disclosure is executed in a computer-readable program product comprising a computer-readable medium. The computer-readable medium may comprise computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology CD-ROM, digital versatile disks ("DVD"), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer.

[00461 The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope and spirit of the disclosed embodiments. It must also be noted that as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise.

[00471 Advantages of the embodiment of the present disclosure are illustrated herein a. Grouping the riders based on social attitude parameters helps provide riders a pleasant riding experience.

100481 Referral Numerals:

Reference Number Description

System 102 Processor 104 Memory 106 Transceiver Matching server 112 Network Rider 122 User equipment 501-507 Method steps

Claims

[00501 Claims: We Claim: 1. A method (500) for forming ride-sharing groups of one or more riders, comprising: receiving (501) a ride sharing request from a rider, wherein the ride sharing request comprises source and destination of the requested ride, and one or more demographic details of the rider; receiving (503) a social attitude parameters of each rider; I 0 determining (505) a context information based on the social attitude parameters of each rider; calculating (507) a score of each rider based on the received ride sharing request and the context information; and predicting (509) a ride-sharing groups comprising at least two riders based on comparison IS of the calculated score of each rider.
2. The method as claimed in claim I, wherein the social attitude parameters of each rider include his personal interest information.
3. The method as claimed in claim 1, wherein the social attitude parameters are determined from one or more sources, wherein the one or more sources include at least one of: social networking profile of the rider; one or more sensors placed inside a vehicle which the rider is riding; reply to a questionnaire provided by the rider; post ride experience shared by the rider; and one or more sensors placed outside the vehicle.
4. The method as claimed in claim 1, wherein the calculated score is expressed in the form of a continuous target variable.
5. A system ( [00) for forming ride-sharing groups of one or more riders, comprising: a memory (104); a transceiver (106) configured to: receive a ride sharing request from each rider, wherein the ride sharing request comprises source and destination of the requested ride of each rider and one or more demographic details of each rider; receive social attitude parameters of each rider; and a processor (102) configured to: determine a context information based on the social attitude parameters of each rider; calculate a score of each rider based on the received ride sharing request and the context information; and predict ride-sharing groups comprising at least two riders based on comparison of the calculated score of each rider.
6. The system as claimed in claim 1, wherein the social attitude parameters of each rider include his personal interest information.
7. The system as claimed in claim 1, wherein the social attitude parameters are determined from one or more sources, wherein the one or more sources include at least one of: social networking profile of the rider; one or more sensors placed inside a vehicle in which the rider is riding; reply to a questionnaire provided by the rider: post ride experience shared by the rider; and one or more sensors placed outside the vehicle.
8. The system as claimed in claim 1, wherein the calculated score is expressed in the form of a continuous target variable.
9. The system as claimed in claim 1, further comprising: a matching server 110 coupled to the system 100.