JP2017513135A - Method for delivering information to meet the current demands of vehicle drivers - Google Patents

Abstract

The method first delivers information to meet the current demands of the driver of the vehicle by obtaining data relating to the current state of the vehicle. The current state and the prediction model are passed to the prediction procedure to determine the node selection probability in the selection tree (CT). Based on the selection probability, information relating to the current request of the driver is selected from a database storing commercial information and non-commercial information. The information is then delivered to the driver. Delivery is autonomous from the driver.

Description

  The present invention relates generally to a method and system for delivering information to a driver of a vehicle, and more particularly to providing precise and useful information related to the current state associated with the driver and / or vehicle. Relating to delivering.

  The main purpose of the driver driving the vehicle is to reach the destination safely in a reasonable amount of time. The second objective can include maintaining the condition of the vehicle and providing comfort and entertainment to the occupant. Achieving these objectives may require a wide variety of information such as information about restaurants, rest facilities, hotels, service stations, entertainment facilities, and the like.

  There are typically multiple options, and the option selected by the driver may depend on cost, quality, location and / or driver preference. Traditionally, information is delivered by commercial shoulder signs or locally broadcast commercial advertisements, but cannot predict any particular driver preference or demand.

  A much better approach is to set delivery goals for specific consumers. Typically, targeted delivery is accomplished by a recommendation system that uses collaborative filtering or content-based filtering such as Google AdWords®, Microsoft Bing® Ads, and the like. On the other hand, these recommendation systems were not designed for in-vehicle use. Conventional systems use a large database of consumer purchase patterns, such as information about previously purchased or used products. Interacting with such online systems can be complex and extremely distracting.

  Traditional information delivery usually focuses only on commercial advertising. In this regard, Patent Document 1 "Apparatus and Method for Mobile Intelligent Advertizing Service Based on Mobile user Contextual Matching", Patent Document 2 "Intelligent Presentation of Advertising with Navigation", Patent Document 3 "Apparatus and Method for Rendering Advertising Contents on a Mobile Communication Device "and Patent Document 4" Method and Apparatus for Selecting Advertisements and Deter " ining Constraints for Presenting the Advertisements on Mobile Communication Devices ", which is incorporated herein by reference.

US Patent Application Publication No. 2012/0101903 US Patent Application Publication No. 2012/0221413 US Patent Application Publication No. 2011/0066501 US Patent Application Publication No. 2009/0298483

  Embodiments of the present invention provide systems and methods for autonomously delivering commercial and non-commercial information to vehicle drivers. The driver can be regarded as a potential consumer of information. In particular, the information relates to goods and services that are of potential interest to the driver. The method identifies potential requests in the form of a probability distribution over the selection tree and determines what information to deliver. The method is based on the recognition that the driver probably wants to consider a relatively small amount of relevant information rather than selecting from a large number of untargeted roadside signs and broadcast commercials.

2 is a flow diagram of a system and method for delivering information to a driver of a vehicle according to an embodiment of the invention. 3 is an exemplary selection tree used by embodiments of the present invention. 4 is a record used by a system and method according to an embodiment.

  Embodiments of the present invention provide systems and methods for autonomously delivering information to meet the current needs of vehicle drivers. It should be noted that the information can also relate to other occupants. However, the main purpose is to eliminate driver input to the system.

  Information differs from commercial (marketing) information, eg, advertisements and prior art systems, such as non-commercial (public service) information, eg, traffic or road conditions, public parking, unadvertised radio stations, etc. Can take form.

  The method identifies a driver's current demand and then autonomously delivers information related to products and services that potentially satisfy the current demand. It should be understood that the information can be any type of information useful to the driver. It should also be understood that commercial and non-commercial information can compete for driver attention. As used herein, the term “autonomous” means that the method does not require direct input from the driver by voice or typed queries, as in conventional general systems and methods. Means.

  The method is based on the recognition that the designation of requirements in terms of typed input or voice input, as in the prior art, is unsafe and / or inaccurate for the driver. The method effectively uses current and past conditions associated with the driver and / or vehicle during a search and selection procedure for products and services that may be of interest to the driver, thereby Relevant information can be delivered autonomously to meet the driver's current requirements.

  For example, current demands are based on drivers frequently visiting local coffee shops operated by specific chains while listening to specific news programs from 8am to 9am on weekdays. there is a possibility. Thus, if the system detects an event that the driver is in a different location during the current state, the system is concerned with coffee shops in the same chain while tuning to the driver's preferred news program at the correct time. Information to be delivered autonomously. The driver only drives. As another example, if the vehicle currently requires fuel or repair, information about the location of the nearest service station of the brand preferred by the driver can be delivered. Since the delivery of this information is of commercial interest to merchandise and service providers identified as useful to the driver, advertising revenue is collected from these providers.

  To do this, the method obtains data 100 as shown in FIG. Data acquisition can be continuous and / or periodic. The data can include data relating to the driver and vehicle, for example, navigation data 101, vehicle bus data 102, for example, fuel level, speed, etc., indoor and outdoor climate data 103, time and date 104, State 110 is formed. The current state can take the form of a vector. The current state is used to update the past state 151 in the database. Updates form a good characterization of the current and past states used to maintain nodes in a choice tree (CT) 200 as described below, as well as observed request selections. Done frequently enough.

  In one embodiment of the invention, as shown in FIG. 1, the method is performed by a system operated by an in-car information communication system (ICICS) 180 that is in wireless communication with a vehicle 190. Is called. That is, the data 100 is transmitted from the vehicle to the ICICS using the wireless communication system. ICICS maintains past states, CTs and prediction models, performs prediction procedures, and sends information to the vehicle based on selection probabilities. There may be one CT per driver and / or vehicle. That is, the identification information of the driver or the vehicle is transmitted to the ICICS together with each current state, so that the ICICS can select an appropriate CT to be operated.

  In another embodiment of the present invention, most of the ICICS functions such as CT maintenance and updating are performed locally in the vehicle, and only the database of available advertisements is continuously updated over the wireless connection.

  The current state and past state are used to maintain the predictive model 160. The prediction model 160 relates the current state and the past state. The prediction model can be learned from the data 100, created manually, or obtained in some other way. For this, see for example related patent applications. The prediction procedure 130 determines a selection possibility 131 for the nodes in the CT using the prediction model and the current state. One example of such a forecasting procedure is U.S. Patent Application No. 14 / 035,502, “Method and System for Dynamically Adapting User Interfaces in Vehicles in Vehicular Science in Vehicles, filed September 24, 2013 by Nikovski et al. This US patent application is hereby incorporated by reference. The possibility of selection is used to autonomously deliver and present specific information to the driver using an output device 170, such as a display screen or loudspeaker. For this, see the related application. In embodiments, autonomous delivery is initiated by the current state or a relatively recent state rather than by a driver interacting with the system as in the prior art.

  The method may be performed on one or more processors connected to memory and input / output interfaces known in the art. The processor can be connected to a transceiver for wireless communication.

  As shown in FIG. 2, possible requests and information about the requests are organized in a selection tree (CT) 200. CTs are organized according to the types of products and services that a driver may need. The top of the tree is a root node 201. The first intermediate level 210 represents general demands such as food, fuel, lodging and entertainment facilities. This level may also include requirements related to other top level requirements of the driver, such as temperature control, map manipulation, etc. The other intermediate level (s) 220 are more specific and specify, for example, general request categories. The lowest (leaf node) level 230 relates to products and services that meet specific requirements. In other words, the node becomes more specific as it goes down the CT level.

  One method for determining user requirements is U.S. Patent Application No. 14 / 035,502 filed by Nikovski et al. "Method and System for Dynamically Adapting User Interfaces in Vehicles in Vehicles in Vehicular Science in Vehicular Science in the United States. "Complexity". This US patent application is hereby incorporated by reference. The method adapts the user interface of the vehicle navigation system based on an input vector that represents the current state associated with the vehicle. In order to achieve the next state using a predictive model representing the previous state, the probability of motion is predicted. Next, the subset of actions with the highest probability that minimize the complexity of the interaction with the vehicle navigation system is displayed in the vehicle.

  US patent application Ser. No. 14 / 075,939, “Method for Predicting Travel Destination Based on Historical Data,” filed by Hershey et al. On November 8, 2013, is incorporated herein by reference. Describes how to determine the visit probability of a location.

  FIG. 3 shows the format of a record 300 associated with a node in the CT. The record is associated with a label 310, a strength value (SV) 320, a cost 330, information 340 about the product or service, eg text, image or eg audio, video or music. And a pointer 350 to the existing CT node. By multiplying the selection probability of the node pointed to by the pointer 350 by SV 320, an estimate of the relevance of the node to meet the driver's current demand is obtained.

Labels Labels identify products or services at various levels of the selected tree, such as food and accommodation at the first level 210, coffee and hotels at the intermediate level 220, and Starbucks and Hilton at the leaf nodes at the lowest level 230 To do. The label is more specific as the CT level is lower.

SV
The SV indicates how much the information 340 is related to the related node. SV represents the probability that the consumer / driver will respond to the information, and this probability should be proportional to the likelihood that the information will result in meeting the specified requirements. For example, information about a gas station has a high intensity value associated with a node in the CT corresponding to the fuel. This is because all gas stations meet fuel requirements. In contrast, gas stations have lower SVs for CT nodes corresponding to food. This is because not all gas stations sell food, and the demand for food is not met in all cases where a driver stops at a gas station. The exact SV value can be adjusted based on data collected during operation. The SV is initially set to a default value and can be updated based on data 100 obtained from many drivers. The selection process uses the current SV to selectively deliver information to the consumer.

  Each time the SV is updated, the method uses the current probability to select information that is preferably delivered. In one embodiment, the selection process delivers information associated with the highest SV corresponding to the leaf node with the highest selection probability. This selection maximizes the likelihood that the driver's primary requirements will be met and the driver will actually use the delivered information. For example, if the system determines that the driver currently needs coffee, the information most relevant to coffee is selected from the entire database.

  In other embodiments, the information delivered to the driver maximizes the product of SV, cost, and the selection probability of one of the nodes whose labels are associated with the information. This delivery strategy is intended to maximize revenue for ICICS.

Cost As used herein, the term “cost” is broadly interpreted as any metric derived from, for example, monetary value, positive or negative incentives, or some cost function. In one embodiment, the cost represents the amount that merchandise and service providers want to pay ICICS to deliver information. Payment terms may be, for example, cost per impression (s) (CPI) or cost per thousand impressions (CPM), i.e., the number of times information is delivered to a consumer, or a "click" cost (CPC). : Cost per “click”), ie, the number of times the information was actually used, but in the system according to embodiments of the present invention there is no actual “click action”. When CPC is used, a coupon code can be delivered along with the information. When a coupon code is used at a provider, the use of the information can be verified and the consumer can be awarded accordingly. ICICS can also send coupon codes to mobile devices such as mobile phones or tablet computers for later use outside the vehicle.

  The method can monetize the improvement in efficiency through auctions where providers bid for higher level labels in the CT, such as “food”, “fuel” or “hotel”. In this type of auction, the cost of information at all levels is determined as the product of the CPC bid price and their relevance.

  Unlike AdWords®, information relevance is estimated as the selection probability for the CT with which the information is associated. In this information selection mechanism, only one CT node is associated with the information, but the same information can be contained in multiple records, each record being associated with a different CT node and possibly a different cost. .

  After the probabilities for all records are determined, the information in the record with the highest selection probability is delivered and the amount the provider wants to pay is the second highest selection probability, the relevance of the highest selection probability Is divided by and added as a small fixed amount, for example, $ 0.01. This mechanism selects information relevant to the consumer and minimizes the cost paid by the provider in non-competing situations. This can be a great incentive for merchandise and service providers to participate in ICICS.

  Another selection process allows the provider to bid on the CT node and, for example, a logical condition regarding the confidence value associated with the information when the information is displayed. Continuing the coffee example above, two coffee providers at the leaf node, such as Starbucks and Dunkin Donuts, may bid $ 0.03 and $ 0.02 at the second level node [coffee]. it can. Furthermore, Dunkin Donuts has a logical condition of 0. 0 when the selection probability of the node [coffee] is higher than 0.7 and the selection probability of the child node [Starbucks] is higher than the selection probability of the sibling node [Dunkin Dunts]. You can bid $ 05. In other words, when a driver's request for his / her product is identified, but the driver may be heading for a competitor, he / she bids further for the privilege of showing his / her information. It can be performed. In that case, different information can be delivered for the purpose of attracting the consumer, for example information that provides a greater discount. If multiple advertisements meet the logical condition, the information in the node with the highest cost can be used or an auction can be conducted.

  Furthermore, when two or more providers bid on the same node, the cost of delivering information can be based on the bid amount and the current location of the vehicle. For example, Starbucks, Dunkin Donuts and Pete's Coffee are bidding $ 0.03, $ 0.02 and $ 0.01 against node “coffee”, but only Dunkin Donuts and Pete's Coffee If this is the preferred choice at the moment, Dunkin Donuts information is delivered instead of Starbucks. This is equivalent to adding another logical condition to the scheme described above so that the condition is true when the provider can be reached within the specified time limit. This time limit can be user specific.

  The SV can be updated for all drivers that interact with the CT. The primary source for positive association is the successful retrieval of the coupon code. The absence of such a collection within a specific period of information presenting information to the driver can be considered as an indicator of a negative association between the node in the CT and the information. Further, the negative association indicator may be more directly due to a “thumbs down” button available with the displayed information. This button allows the driver to reject the delivered product or service. The corresponding SV can be greatly reduced, so that this information is less likely to be delivered when the CT node selection probability increases.

  The method can also be used for travel planning and is filed by Harsham et al. On Mar. 6, 2014, US patent application Ser. No. 14 / 198,742, “Actions Prediction for Hypothetical Driving Conditions”. ”Describes a prediction engine for predicting behavior based on a set of driving state parameters and a driving history, and a simulation engine for generating a virtual scenario corresponding to the traveling. After the driver has selected the starting point and destination of the journey, along with the date and time of the journey, the journey planning tool uses the method according to the present invention to identify and satisfy this demand. A list of possible information can be prepared along with a set of personal coupons for the driver to use during future driving. Perhaps such a personal coupon set would be more valuable and effective than using a general coupon book distributed by a local tourism association.

Claims (19)

A method of delivering information to meet the current demands of a vehicle driver,
Obtaining data relating to the current state of the vehicle;
Passing the current state and prediction model to a prediction procedure, determining a selection probability of a node in a selection tree (CT);
Selecting information related to the current request of the driver from a database storing commercial information and non-commercial information based on the selection probability;
Delivering the information to the driver, wherein the delivering is autonomous from the driver;
And the method is performed on one or more processors. The information is associated with the node having the highest selection probability;
The method of claim 1. The obtaining, passing, selecting and delivering are performed by an in-vehicle information communication system (ICICS) connected to the vehicle by a wireless communication system.
The method of claim 1. The information includes commercial information in the form of advertisements for products and services, and vendors of the products and services pay the ICICS operator to deliver the information to the driver.
The method of claim 3. The information includes non-commercial information,
The method of claim 1. The selection probability identifies the driver's current request;
The method of claim 1. The information is presented to the driver at an output device;
The method of claim 1. The obtaining, updating, passing, selecting and delivering is initiated only by the current state or a relatively recent state.
The method of claim 1. The nodes are organized in the CT according to the types of products and services to meet the requirements, and the CTs are root nodes, general request intermediate level nodes, and products and services that satisfy specific requirements. Including related leaf nodes,
The method of claim 1. The database includes records, each record including a label, a pointer to an associated node in the CT, an association strength value (SV) to the node, a cost, and the information.
The method of claim 1. Further comprising multiplying the selection probability by the SV to obtain an estimate of the relevance of the node to satisfy the driver's current request;
The method of claim 10. The SV represents the probability that the driver will respond to the information.
The method of claim 10. The information delivered to the driver maximizes the product of the SV, the cost, and the selection probability so as to maximize revenue for an in-vehicle information communication system (ICICS);
The method of claim 10. The cost represents the amount that goods and service providers want to pay to the ICICS to deliver the information;
The method of claim 13. Financial terms are based on cost-per-thousand impressions,
The method according to claim 14. Further comprising delivering a coupon code with the information to prove use of the information;
The method of claim 1. The cost is determined by auction,
The method of claim 10. Further comprising planning a run using the method;
The method of claim 1. The information is associated with a logical condition when the information is displayed;
The method of claim 1.

Images