Browse Prior Art Database

Method and Apparatus of Experience-aware Effective Trajectory Pattern Detection for Route Recommendation

IP.com Disclosure Number: IPCOM000239606D
Publication Date: 2014-Nov-19
Document File: 4 page(s) / 73K

Publishing Venue

The IP.com Prior Art Database

Abstract

Route recommendation is an important issue for smart traffic and connected vehicle solutions. Existing route recommendation methods haven't taken individuals traveling experience into consideration. This disclosure proposes a method to generalize effective route patterns from individuals travel experience. The core idea is to identify the most experienced user for each origin-destination by driving experience efficiency assessment, then extract the most effective route took by this experienced user under different context: 1) Driving experience efficiency assessment. Assessing the user’s driving experience from two aspects: efficiency assessment for different users in the same OD and efficiency assessment for the same user in different ODs. The former is to compare the experience efficiency for different users which is evaluated thru metrics like frequency and cost time. The latter is to compare the route familiarity in different OD for the same user and filter the user with bad driving behaviors (such as over speed); 2) Context aware route recommendation. In order to extract the most efficient route under different context for each OD, the route history for the most experienced user needs to be clustered into different context category firstly. Then, the recommended route will be the route with the most high rating within each context category. Compared with the existing solutions, our invention is more effective because the route we recommend is leveraging the human insight of the most experienced driver with good driving behavior and the route we recommended is suitable for different context. we claim a method and system of experience-aware effective trajectory pattern detection for route recommendation, which leverages the human insight of the driver who is most familiar with the route and also with good driving behavior, and finally recommends the route for the query considering the different context. We claim a method and system which leverages human experiences to recommend the effective route for vehicle drivers. Vehicle drivers have intelligent insight in their familiar origin-destination (such as from home to office). This kind of human insight is effective but complex to represent. Thru assessing driving experience efficiency, the most efficient route took by experienced user can be recommended to other users for specific origin-destination.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 4

Method and Apparatus of Experience

Method and Apparatus of Experience-

Our invention is to solve the problem of detecting trajectory pattern for route recommendation in the domain of connected vehicle .

The known solutions for this problem include:

1) Shortest distance based recommendation: this method recommends the route with the shortest distance in the route network. However, this method only based on the static route network that is not suitable for different context (such as time, traffic);


2) Minimum duration based recommendation: this method recommends the route with the minimum duration of the trajectory. However, this method is not effective because the trajectory with minimum duration sometimes is produced by the driver with bad driving behavior (such as over speed, run the red light) and is also not suitable for different context;


3) Frequency based popular route recommendation: this method recommends the route with the most intensive trajectories. This method leverages the humans' frequency trajectory but not all humans with good insight on the effective route that simply based on the statistical result is not effective sometimes. And also, this method does not consider the different context.

Here, we claim a method and system of experience-aware effective trajectory pattern detection for route recommendation, which leverages the human insight of the driver who is most familiar with the route and also with good driving behavior, and finally recommends the route for the query considering the different context.

We claim a method and system which leverages human experiences to recommend the effective route for vehicle drivers. Vehicle drivers have intelligent insight in their familiar origin-destination (such as from home to office). This ki...