Map Generation Device Using Center Point Sequences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map generation methods using probe data reflect biased information due to varying traveling states of vehicles, affecting accuracy, especially between high-speed and low-speed travels.
Innovation Solution
A map generation device categorizes shape point sequences of marking lines into groups for vehicles, generates center point sequences to remove state-dependent information, and identifies these sequences as marking lines on a map, thereby generating accurate map data independent of vehicle traveling states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If shape point sequences from multiple vehicles are directly used to generate map data, then map generation can be performed using available probe data, but the map accuracy deteriorates due to inclusion of state-dependent information from different traveling states
Solution Approach 1:
The patent segments the shape point sequences by grouping them according to vehicle traveling states (high-speed, low-speed, etc.). This segmentation allows separate processing of sequences from different traveling states, enabling the system to handle and neutralize state-dependent information systematically, thereby improving map generation accuracy.
Solution Approach 2:
The patent changes the parameter representation by converting shape point sequences into center point sequences. This parameter transformation removes state-dependent information while preserving the essential geometric characteristics of marking lines, thus improving map accuracy without losing critical spatial information.
2Measurement precision
If center point sequences are generated to remove state-dependent information, then map accuracy is improved, but the processing complexity increases due to additional grouping and sequence generation steps
Solution Approach 1:
By segmenting shape point sequences into groups based on traveling states, the patent enables systematic processing that, while adding a grouping step, simplifies the overall logic by handling each group uniformly. This structured approach manages complexity through organization rather than ad-hoc processing.
Solution Approach 2:
The transformation from shape point sequences to center point sequences simplifies the data representation by reducing dimensionality and removing state-dependent variables. This parameter change, while adding a processing step, ultimately reduces complexity by creating a more manageable and uniform data structure for map generation.
3Productivity
If shape point sequences are processed without grouping by vehicle, then processing speed is maintained, but map accuracy deteriorates due to mixed state-dependent information
Solution Approach 1:
The patent segments processing into grouping and center point generation phases. While this adds a preliminary step, it enables highly efficient uniform processing of each group, maintaining overall productivity. The segmentation prevents expensive corrective processing later by addressing state-dependency early in the workflow.
Solution Approach 2:
By performing the parameter transformation to center point sequences in a systematic grouped manner, the patent achieves efficient batch processing. This approach maintains productivity by processing multiple sequences uniformly while simultaneously improving accuracy through consistent state-independent representation.
Data Source
AI summary
A server includes a probe data acquiring unit acquiring a probe data item of each of vehicles traveling on a road, the probe data items include, as shape point sequences, sequences of shape points of marking lines; a grouping unit categorizing shape point sequences of marking lines into plural groups for the respective vehicles, each group including one or more shape point sequences for a corresponding one of the vehicles; a center point sequence generating unit generating, based on the one or more shape point sequences included in each group, a center point sequence for each marking line to thereby remove, from the shape point sequences of the marking lines, information which is dependent on traveling states of the vehicles; and a map data generating unit identifying the center point sequence for each marking line as a corresponding marking line on a map to thereby generate map data.