Crowdsourced HAD Map Updates via Multi-Terminal 3D Reconstruction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for updating Highly Automated Driving (HAD) maps are economically unviable due to the high cost of dispatching survey vehicles for minor changes and the quality degradation from using low-resolution data from low-cost equipment.

Innovation Solution

A system and method for updating HAD maps through crowdsourcing data from multiple terminals, where mobile terminals capture images from varying view positions and a server integrates this data to reconstruct 3D models and update the maps, reducing the need for frequent resurveying by high-cost vehicles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If survey vehicles are dispatched to update HAD maps, then map update accuracy is maintained, but operational costs increase significantly

Engineering Contradiction:
Improvemap update accuracyVSAvoidoperational cost
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent uses mobile terminals (smartphones, tablets) as inexpensive copies of expensive survey vehicles to capture map data. These terminals replicate the data collection function without requiring specialized equipment, thereby maintaining measurement capability while dramatically reducing operational costs.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system enables mobile terminals to perform multiple functions: capturing images, detecting changes, and contributing to map updates. By making ordinary devices multi-functional, the system eliminates the need for dedicated survey vehicles while maintaining map update accuracy.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If low-cost equipment is used to update HAD maps, then operational costs decrease, but data quality and map resolution deteriorate

Engineering Contradiction:
Improveoperational costVSAvoiddata quality
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent merges data from multiple mobile terminals through a server platform. By combining images and detection results from numerous low-cost devices, the system achieves data quality comparable to expensive survey vehicles while maintaining low operational costs.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system implements feedback mechanisms where the server analyzes incoming data quality, requests additional images or corrections from terminals, and validates detection results. This feedback loop ensures that low-cost equipment produces high-quality map updates through iterative refinement.

Inventive Principle:
Principle #23Feedback

3Productivity

If multiple mobile terminals are used for data collection, then operational cost decreases, but data integration complexity increases

Engineering Contradiction:
Improveoperational costVSAvoiddata integration complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces a server as an intermediary between mobile terminals and the map database. This intermediary handles all data integration, processing, and validation tasks, shielding the complexity from both the terminals and the final map update process. The server standardizes data formats and coordinates terminal activities.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3610225B1Systems and methods for updating highly automated driving maps
Publication Date: 2022.03.02 BEIJING DIDI INFINITY TECH & DEV CO LTD
  • EP3610225B1 patent drawingFigure 1
  • EP3610225B1 patent drawingFigure 2
  • EP3610225B1 patent drawingFigure 3

AI summary

A system (100) for updating a HAD map using a plurality of terminals (131-134). The system (100) may include a communication interface (202) configured to communicate with the plurality of terminals (131-134) via a network, a storage (214) configured to store the HAD map, at least one processor (204). At least one processor (204) may be configured to identify a target region including at least one changing object, instruct the plurality of terminals (131-134) to acquire data of the target region at varying view positions, and receive the acquired data from the plurality of terminals (131-134). The at least one processor (204) may be further configured to construct a three-dimensional model for the at least one changing object from the acquired data, and update the HAD map based on the three-dimensional model. Also provide a method for updating a HAD map using a plurality of terminals (131-134) and a non-transitory computer-readable medium having a computer program stored thereon. This provide an improved system and method for updating a high-resolution map.