Roadside Perception Filtering for Autonomous Vehicle Path Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automatic driving vehicles rely on roadside devices for perception information, but due to limited detection ranges and potential obstructions, they often default to using incomplete perception data, leading to safety hazards as they may overlook obstacles in unperceived areas.
Innovation Solution
The roadside device filters and sends targeted road perception information along with effective area data to the vehicle, allowing it to accurately navigate based on relevant information, reducing the risk of accidents by avoiding the use of incomplete data and minimizing unnecessary data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the roadside device sends all detected road perception information to the automatic driving vehicle, then the vehicle can access comprehensive environmental data, but the data transmission amount increases and includes ineffective information from areas outside the vehicle's actual path
Solution Approach 1:
The roadside device extracts only the road perception information from the effective area (where the vehicle is actually traveling) and sends it to the vehicle, separating useful information from unnecessary data. This extraction principle reduces data transmission amount while maintaining information completeness for the vehicle's actual path.
Solution Approach 2:
The patent applies local quality by differentiating between effective area information (relevant to vehicle path) and non-effective area information (irrelevant to vehicle path). The roadside device identifies and transmits only the local portion of perception data that corresponds to the vehicle's actual traveling area, making the data transmission targeted and efficient.
2Ease of operation
If the vehicle uses road perception information from the entire coverage area by default, then it can maintain simple processing logic, but it may use incomplete perception data from areas where detection is obstructed or ineffective
Solution Approach 1:
The roadside device performs preliminary action by pre-identifying and marking the effective area (where perception is reliable) before transmitting data to the vehicle. This preliminary classification allows the vehicle to simply use the provided information without complex validation logic, while ensuring reliability through advance filtering of effective perception data.
Solution Approach 2:
The roadside device acts as an intermediary that filters and validates perception information before transmitting it to the vehicle. It mediates between the complex detection environment and the vehicle's processing system, providing only reliable data from effective areas, thus maintaining both simplicity for the vehicle and high reliability of the information.
3Device complexity
If the roadside device transmits perception information for the entire coverage area, then no additional filtering processing is needed, but the vehicle receives ineffective information from areas outside its actual path
Solution Approach 1:
The roadside device applies local quality by identifying and transmitting only the perception information from the effective area (where the vehicle is actually traveling). This localized approach filters out irrelevant information from areas outside the vehicle's path, maintaining information relevance without requiring complex filtering algorithms.
Solution Approach 2:
The patent changes the parameter of information relevance by using effective area identification to filter perception data. The roadside device transmits information with an associated有效性 flag or spatial reference, allowing the vehicle to quickly identify relevant data without complex filtering, thus reducing processing complexity while maintaining high relevance.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The present application discloses a vehicle control method, a vehicle-road coordination system, a roadside device and an automatic driving vehicle, and relates to the technical fields of artificial intelligence, automatic driving, intelligent transportation, and computer vision. The specific solution is: when an automatic driving vehicle is controlled to travel, road perception information acquired in a coverage area of a roadside device is firstly filtered; effective area information of an area to which target road perception information obtained by filtering belongs is determined; then the target road perception information, the effective area information and area information of the coverage area of the roadside device are sent to the vehicle together, so that the vehicle can accurately control the vehicle to travel in combination with the target road perception information, the effective area information and the area information, hence improving the traveling safety of the vehicle; in addition, the target road perception information that has an effective reference value on the traveling of the vehicle is sent to the vehicle in a targeted manner, thereby reducing a transmission amount of data.