Roadway Object Deduplication for Crowdsourced Vehicle Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle systems face inefficiencies due to duplicate object clusters and contradictions in crowdsourced data, leading to computing resource waste and reduced confidence in map generation and vehicle control.
Innovation Solution
A vehicle system with a control module that clusters data from multiple vehicles, identifies and merges or removes duplicate groups based on proximity, sampling frequency, and weighted values, and resolves contradictions using transportation regulations and historical data to determine accurate object locations and values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If duplicate object clusters are not resolved, then more data points are available for analysis, but computing resources are wasted and map generation precision is reduced
Solution Approach 1:
The system merges duplicate object clusters by identifying clusters with overlapping spatial boundaries and combining their data points into a single unified cluster. This reduces redundant computations while preserving comprehensive data coverage, directly resolving the contradiction between maintaining data completeness and improving computing efficiency.
Solution Approach 2:
The system extracts and removes duplicate clusters from the data set by identifying and eliminating redundant spatial groups. This extraction process maintains only unique object representations while preserving the essential data points, thereby improving map generation precision without losing critical information.
2Reliability
If all detected object locations are included without deduplication, then data completeness is maintained, but contradictions in data quality increase and confidence in vehicle control is reduced
Solution Approach 1:
The system applies different quality assessment criteria to different spatial regions and object types. By evaluating local data characteristics such as detection frequency, spatial consistency, and contextual relevance, the system selectively includes or excludes data points to maintain overall reliability while preserving locally important information.
Solution Approach 2:
The system implements a feedback mechanism that continuously monitors data quality metrics and adjusts the inclusion of data points accordingly. When contradictions are detected in data quality, the system uses feedback loops to resolve ambiguities and maintain confidence in vehicle control decisions while preserving essential data for map generation.
Data Source
AI summary
A vehicle system includes a vehicle control module of a vehicle and a control module. The control module is configured to receive data from vehicles over a period of time, cluster the received data into groups, detect at least one set of duplicate groups of the groups, merge the duplicate groups into a single group or remove at least one of the duplicate groups to form a set of deduplicated groups associated with the at least one object, and determine an expected location of the at least one object based on the set of deduplicated groups. The vehicle control module is configured to receive the expected location of the at least one object and generate a control signal for controlling at least one operation of the vehicle based on the expected location of the at least one object. Other example systems and methods are also disclosed.