Traffic Object Map Updating From Multi-Vehicle Detection Discrepancies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in updating their internal maps to reflect changes in the environment, such as traffic object positions and types, which can lead to navigation errors due to discrepancies like a stop sign being replaced by a traffic signal during construction, resulting in potential collisions or unsafe maneuvers.
Innovation Solution
A sensor positioning platform that dynamically repositions and rotates sensors to enhance coverage and accuracy, allowing for real-time updates of traffic maps by detecting changes in traffic objects, utilizing a combination of camera data and machine learning for accurate object detection and map updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are mounted at fixed locations on autonomous vehicles, then the device complexity is reduced and ease of operation is improved, but the measurement precision and reliability of traffic map updates deteriorate due to limited coverage and inability to adapt to environmental changes
Solution Approach 1:
The patent applies the Dynamics principle by transitioning from fixed sensor mounting to dynamic sensor positioning. The sensor positioning platform can rotate and reposition sensors to different locations and orientations, allowing the system to adapt to environmental changes and improve detection accuracy. This dynamic capability enables the sensors to capture traffic objects from multiple angles and positions, resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The sensor positioning platform serves multiple functions: it can rotate sensors to different orientations, reposition them to various locations, and adjust their viewing angles. This multi-functionality allows a single platform to replace multiple fixed sensor installations, improving measurement precision while managing device complexity through consolidation of functions into one versatile system.
2Reliability
If internal maps are not updated with real-time sensor data, then the loss of information is minimized and stability is maintained, but the reliability of autonomous vehicle navigation deteriorates due to discrepancies between map data and actual environmental conditions
Solution Approach 1:
The patent implements feedback by continuously comparing sensor-detected traffic objects with existing map data and updating the maps when discrepancies are found. The system detects traffic objects such as traffic lights, signs, and signals, compares their positions and characteristics against the internal map, and initiates map updates when differences exceed a threshold. This feedback loop ensures navigation reliability by keeping map data current while filtering out minor variations.
Solution Approach 2:
The system performs preliminary comparison of sensor data with map data before committing to updates. By pre-processing and validating detected objects against existing map information, the system can identify significant changes that require map updates while ignoring normal variations, thus maintaining reliability without excessive information loss.
3Measurement precision
If sensor coverage is expanded to detect all traffic objects, then the measurement precision and reliability improve, but the device complexity and cost increase due to additional sensors and processing requirements
Solution Approach 1:
Rather than deploying multiple fixed sensors simultaneously, the patent uses a single sensor mounted on a dynamic positioning platform that rotates and repositions to achieve comprehensive coverage. This dynamic approach provides equivalent or superior detection capability compared to multiple fixed sensors, improving measurement precision while reducing device complexity by using one versatile sensor instead of many static ones.
Data Source
AI summary
Systems, methods, and computer-readable media are provided for receiving traffic object data from a plurality of autonomous vehicles, comparing the traffic object data of each of the plurality of autonomous vehicles with known traffic object data, determining a discrepancy between the traffic object data of each of the plurality of autonomous vehicles and the known traffic object data, grouping the traffic object data of each of the plurality of autonomous vehicles based on the determining of the discrepancy between the traffic object data of each of the plurality of autonomous vehicles and the known traffic object data, determining whether a group of traffic object data of the grouping of the traffic object data of each of the plurality of autonomous vehicles exceeds a threshold, and updating a traffic object map based on the traffic object data of the group that exceeds the threshold.