Autonomous Vehicle Inspection Split Between Server and Onboard Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle management systems face challenges in efficiently managing autonomous driving vehicles, particularly in reducing the processing load and communication costs on servers while ensuring early detection of vehicle abnormalities, even for minor issues.
Innovation Solution
A vehicle management system that includes a server with a first inspection device and vehicles equipped with control devices, second inspection devices, sensors, and communication devices. The system allows for sequential transmission of vehicle states during autonomous driving, enabling the server to perform first inspections and the vehicles to perform second inspections, with results transmitted only when failures are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the vehicle transmits all inspection data to the server, then the server can accurately grasp the vehicle state, but the processing load and communication cost on the server increase
Solution Approach 1:
The inspection items are divided into two categories: first inspection items (transmitted to server) and second inspection items (processed locally). This segmentation allows the system to balance between comprehensive monitoring and server load reduction by selectively transmitting only certain inspection results.
Solution Approach 2:
Different inspection items are handled with different processing qualities - some items are fully transmitted to the server for centralized analysis, while others are processed locally with only failure results transmitted. This local quality differentiation optimizes the balance between monitoring accuracy and resource consumption.
2Measurement precision
If the vehicle transmits all inspection data to the server, then the server can accurately grasp the vehicle state, but the communication cost increases
Solution Approach 1:
The system extracts only the essential information (failure results of second inspection items) for transmission to the server, leaving out the detailed data that can be processed locally. This extraction principle reduces communication volume while maintaining the ability to detect and report abnormalities.
Solution Approach 2:
Instead of transmitting all inspection data, the system transmits only the necessary portion (failure results), applying partial action to reduce communication overhead while still achieving the goal of abnormality detection.
3Ease of operation
If the server executes all inspections, then centralized control is achieved, but the response time for detecting abnormalities increases
Solution Approach 1:
The second inspection device performs preliminary inspection and filtering locally before transmitting results to the server. This preliminary action at the vehicle end reduces the time required for server processing by pre-screening data and only transmitting when necessary.
Solution Approach 2:
The system dynamically assigns different inspection tasks to different locations (server vs. vehicle) based on the nature of the inspection item. Time-critical inspections are handled locally with immediate detection, while other inspections are centralized at the server.
Data Source
AI summary
A vehicle inspection method includes first and second transmission steps, and first and second determination steps. In the first transmission step, a vehicle transmits a first state detected by a first sensor mounted on the vehicle to a server, during autonomous driving. In the first determination step, the server determines a pass or a fail of the vehicle for the first inspection item by using the first state received from the vehicle. In the second determination step, the vehicle determines a pass or a fail for the second inspection item by using a second state detected by a second sensor mounted on the vehicle, during autonomous driving. In the second transmission step, the vehicle transmits a determination result of the second inspection item to the server in a case where a fail is determined for the second inspection item.