TurnsMap Left-Turn Detection via Mobile Crowdsensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems lack the ability to efficiently and widely distribute information about left-turn protection settings at intersections, leading to increased risk of accidents due to the fragmented and non-digitized nature of traffic signal data, and the high cost of traditional road survey methods.
Innovation Solution
A system called TurnsMap that utilizes mobile crowdsensing data from smartphones to classify left turns as protected or unprotected by analyzing gyroscope and GPS data, employing machine learning and data mining techniques to differentiate left-turn settings and create a comprehensive database for navigation systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road survey vehicles are used to extract traffic light information, then measurement precision is improved, but productivity deteriorates due to prohibitive costs and slow updating speed
Solution Approach 1:
The patent employs crowdsourced mobile devices to automatically collect and upload traffic light data during normal driving operations. The system leverages existing smartphone sensors (GPS, accelerometer, gyroscope, camera) to self-service the data collection function, eliminating the need for dedicated survey vehicles and enabling continuous, cost-effective updating of traffic information across the entire road network.
Solution Approach 2:
The patent makes mobile devices perform multiple functions: navigation, communication, and traffic light detection. By utilizing the universal capabilities of smartphones already carried by drivers, the system avoids the need for specialized equipment, thereby improving productivity while maintaining measurement precision through multi-sensor fusion and machine learning verification.
2Productivity
If mobile crowdsensing is used to collect traffic information, then productivity is improved through widespread data collection, but measurement precision deteriorates due to diverse and noisy sensor data
Solution Approach 1:
The patent merges data from multiple sensors (GPS location, accelerometer motion patterns, gyroscope orientation, and camera images) to compensate for the weaknesses of individual sensors. By combining these diverse data sources through sensor fusion and cross-validation, the system maintains high measurement precision while leveraging the productivity benefits of widespread mobile device deployment.
Solution Approach 2:
The system implements feedback mechanisms where collected traffic light data is verified through machine learning models that compare sensor patterns against expected traffic light behaviors. Inconsistent or noisy data is identified and corrected through this feedback loop, ensuring that the large volume of crowdsourced data maintains high precision despite the diverse and noisy nature of individual sensor readings.
3Reliability
If left turn protection information is made available to navigation systems, then safety is improved by enabling route planning avoidance, but device complexity increases due to fragmented and non-digitized traffic signal data
Solution Approach 1:
The patent replaces the mechanical/manual system of traffic signal specification creation and distribution with an automated electronic system. Mobile devices automatically detect, digitize, and upload traffic light information to a centralized database that navigation systems can access via API, eliminating the need for manual data entry and distribution while reducing infrastructure complexity.
Solution Approach 2:
The patent introduces a centralized database and API service as an intermediary between the fragmented traffic signal data sources and navigation systems. This intermediary layer standardizes data formats, handles data quality issues, and provides unified access to navigation applications, thereby reducing the complexity burden on both data collectors and end-use systems while improving safety through consistent information availability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables widespread and timely updates of left-turn safety information, enhancing navigation systems for both human and autonomous vehicles by identifying safer routes with fewer high-risk left turns, thus reducing accident risks.
Implementation Method 1
a gyroscope of the mobile device while the vehicle is moving
Implementation Method 2
an accelerometer of the mobile device while the vehicle is moving
Implementation Method 3
a global positioning system of the mobile device
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
Left turns are known to be one of the most dangerous driving maneuvers. An effective way to mitigate this safety risk is to install a left-turn enforcement — for example, a protected left-turn signal or all-way stop signs — at every turn that preserves a traffic phase exclusively for left turns. Although this protection scheme can significantly increase the driving safety, information on whether or not a road segment (e.g., intersection) has such a setting is not yet available to the public and navigation systems. This disclosure presents a system that exploits mobile crowdsensing and deep learning to classify the protection settings of left turns.