Map Trajectory Estimation Using Neural Path Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing offline map matching methods require handcrafted features and supervised training, making them inefficient for estimating trajectories of moving objects over extended periods with low sampling rates.
Innovation Solution
An end-to-end deep neural network approach that processes map tile images to estimate trajectories without pre-defined features, using convolutional neural networks and conditional random fields to determine the most probable sequence of candidate paths between successive positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If handcrafted features and supervised training are used for offline map matching, then the method can estimate trajectories of moving objects, but the computational complexity and time consumption increase significantly
Solution Approach 1:
The patent pre-generates all possible candidate paths between road segments and pre-computes their images before receiving trace data. This preliminary preparation allows the neural network to directly score pre-processed candidate paths during trajectory estimation, avoiding real-time feature extraction and significantly reducing computational time while maintaining accuracy
Solution Approach 2:
The patent divides the map into discrete tiles and segments candidate paths into individual image representations. Each candidate path between road segments is converted to a separate image that can be independently scored by the neural network, enabling efficient parallel processing and reducing overall computational complexity
2Ease of manufacture
If handcrafted features are used for map matching, then the processing can be performed, but the method requires feature engineering and supervised training which increases device complexity
Solution Approach 1:
The patent replaces traditional mechanical feature engineering and supervised training processes with a deep learning-based automatic feature learning system. The neural network automatically learns relevant features from raw map tile images and candidate path representations, eliminating the need for manual feature design and reducing system complexity
Solution Approach 2:
The neural network performs self-learning by automatically extracting features from map images and candidate paths without requiring external feature engineering. The system serves itself by learning optimal features during training and applying them automatically during inference, simplifying the overall implementation
Data Source
AI summary
A method is disclosed for estimating a trajectory of an object on a map given a sequence of traces for the moving object. Each trace of the object including information defining a position measured at a given time for the object, as well as information as to an area of accuracy around the measured position. The method processes pairs of successive traces, corresponding to two positions successive in time in the sequence of measured positions for the moving object. For each trace of a pair of successive traces, the method defines road segments on the map within the area of accuracy of the trace. For each road segment within the area of accuracy of a first trace of a pair of traces and each road segment within the area of accuracy of the second trace of the pair, the method determines at least one candidate path between the two road segments. A neural network and a neural graph model are used to compute the most probable sequence of candidate paths to estimate the trajectory of the object on the map.