Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Smart Car - Smart GPRS

IP.com Disclosure Number: IPCOM000250499D
Publication Date: 2017-Jul-26
Document File: 4 page(s) / 20K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 33% of the total text.

Smart Car - Smart GPRS


Disclosed is an intelligent mechanism to predict the destination (or destinations) of a passenger (or a group of passengers in the same vehicle) automatically; by using several inputs from social and digital means, and by processing these using existing cognitive and deep learning capabilities. This prediction is then used to suggest the best route (and / or the best sequence of destinations) to the passenger based on the current traffic and other conditions.

Background of the Problem

There are many GPS apps and driving assistance available like Google maps which are able to detect optimum route based on traffic conditions and other factors (like road blockage etc). Many of these apps also provide this based on auto-prediction of the destination i.e. even without the need of the user providing the destination or opening the app explicitly.

However, there are certain inhibitors

1 - These apps available currently operate either on a pre-defined destination provided by the user, list of favorite destination provided, or based on historical pattern of the user. This limits the capability of the apps to predict the destination accurately only if the user is following the pattern as determined by the app.

2. . Most individuals do not open their GPS applications if they are travelling on  a known route or known destination (not necessarily a frequented destination). This may be due to various factors like battery usage optimization (in case of mobile), confidence on the route, more of disturbance than an aid during known routes etc. 

2. The users may have more than 1 destination, and many times the order of the destination does not matter. For instance if one has to go to his brother and sister’s place on same day, he / she can go to either place first, followed by the second. Is such case, existing systems are unable to suggest the best option.

In this paper, we discuss a mechanism to predict the destination based on several social and digital input parameters and hence detect the destination more accurately even when the journey is not as per the historical pattern or a more frequented destination.

Brief about the solution:

The system achieves automatic detection of destination of a journey by capturing various inputs using existing means. The various inputs and analytics done include

a.    Identifying the passengers in the vehicle (not only the driver but all the drivers)

b.    Capturing the past travel route history of these passengers (driver included)

c.    Social analytics of all passenger's social activity

d.    External information and factors (social gatherings, festival, sports or cultural events like Cricket match or Social gathering etc);

e.    Stored data of the destinations with qualification – example relatives' houses

f.     Knowledge of relationships with the user (like going to sister's house on Rakhsha bandhan festival in India or parents' house on Diwali);

g.    Voice analytics on conversations between the users; or...