Automated Driving Trajectory Divergence for Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated driving vehicles face challenges in maintaining smooth operation due to unexpected obstacles, as existing techniques fail to accurately identify points where they cannot pass smoothly, affecting running stability and safety.
Innovation Solution
An information processing device acquires and compares vehicle-mounted videos from automated and manual driving vehicles to identify divergent trajectories, determining the presence of obstacles and assigning scores to enhance running stability by generating maps and adjusting routes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated driving vehicles rely on existing detection techniques, then the system complexity is reduced, but the measurement precision of obstacle detection deteriorates
Solution Approach 1:
The patent merges video data from multiple vehicles (automated and manual driving vehicles) to collectively identify obstacles. By combining detection results from different vehicles passing through the same road section, the system achieves higher detection precision without each individual vehicle needing complex detection equipment.
Solution Approach 2:
The system uses vehicle-mounted cameras originally designed for general purposes and repurposes them for obstacle detection by comparing trajectories. This multi-functional approach allows existing devices to serve detection purposes, avoiding the need for specialized complex detection systems while maintaining high precision.
2Reliability
If automated driving vehicles compare trajectories with manual driving vehicles, then the reliability of obstacle identification is improved, but the loss of time for data processing increases
Solution Approach 1:
The system performs preliminary trajectory recording during normal vehicle operation, storing trajectory data as vehicles pass through road sections. This preliminary data collection allows for rapid obstacle identification later by simply comparing pre-recorded trajectories, reducing the time needed for actual obstacle detection and verification.
3Measurement precision
If the system identifies divergent trajectories to detect obstacles, then the measurement precision of obstacle location is improved, but the device complexity for trajectory analysis increases
Solution Approach 1:
The system extracts only the essential trajectory information (position coordinates over time) from video data, separating the critical detection elements from the full video content. This extraction approach enables precise obstacle location identification through simple coordinate comparison, avoiding the need for complex video analysis systems.
Data Source
AI summary
An information processing device comprises a controller configured to execute: acquiring respective vehicle-mounted videos including an identical road section from a first vehicle that is an automated driving vehicle and a second vehicle that is a manual driving vehicle; identifying a first running trajectory corresponding to the first vehicle and a second running trajectory corresponding to the second vehicle on the basis of the vehicle-mounted videos; and making a judgment on a subject present at a first point that is a point with a divergence above a predetermined value between the first running trajectory and the second running trajectory.