Autonomous Vehicle Perception Adjustment via Surrounding Vehicle Behavior Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face limitations in perception, leading to inaccurate detection of obstacles and potential false alarms, which can result in emergency stops or other errors due to their reliance on current sensor technologies with limited range and accuracy.
Innovation Solution
A system that analyzes the driving behaviors of surrounding vehicles to improve perception by adjusting and correcting path planning, using both sensor data and behavioral analysis to detect or confirm obstacles, and modify path costs accordingly, potentially adding or removing virtual obstacles based on observed vehicle movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous driving vehicles rely on current sensor technologies for obstacle detection, then the vehicle can operate autonomously with minimal human interaction, but the perception accuracy is limited and false obstacle detections occur frequently
Solution Approach 1:
The patent combines sensor data from the autonomous vehicle with perception data from surrounding vehicles to create a merged perception dataset. This fusion of multiple data sources improves obstacle detection accuracy by cross-validating detections and filling in blind spots, directly addressing the measurement precision limitation while maintaining autonomous operation.
Solution Approach 2:
The system implements feedback by using detection results from surrounding vehicles to correct and refine the autonomous vehicle's own sensor readings. When surrounding vehicles detect obstacles that the ADV's sensors miss or falsely identify, this feedback information is used to adjust the perception model, improving accuracy over time while preserving automation.
2Device complexity
If the autonomous driving vehicle uses sensor data alone for path planning, then the system complexity remains manageable, but the perception range is limited to a certain distance and false alarms increase
Solution Approach 1:
The patent uses surrounding vehicles as intermediaries to extend the autonomous vehicle's perception range. These vehicles act as mobile sensors that relay information about obstacles and road conditions beyond the ADV's own sensor reach, effectively expanding perception coverage without requiring the ADV to carry additional sensors.
Solution Approach 2:
The system makes surrounding vehicles serve multiple functions: they are both traffic participants and extended sensing platforms. By leveraging the perception capabilities of these vehicles for the ADV's path planning, the system gains extended range and reduced false alarms while keeping the ADV's own system complexity manageable.
3Measurement precision
If the autonomous driving vehicle detects obstacles within minimum height requirements, then small obstacles can be detected, but false obstacle detections occur including incorrect location, size, and type identification
Solution Approach 1:
The system uses detection feedback from surrounding vehicles to validate and correct obstacle characteristics. When multiple vehicles detect the same obstacle, their combined data provides feedback that confirms location, size, and type, filtering out false detections and improving both precision and reliability of obstacle identification.
Solution Approach 2:
The patent merges detection data from multiple vehicle sources to create a consolidated obstacle profile. By combining observations of location, size, and type from surrounding vehicles with the ADV's own sensors, the system achieves more reliable and accurate obstacle characterization, reducing false alarms while maintaining detection capability.
Data Source
AI summary
A real-time perception adjustment and correction and driving adaption method for autonomous driving is provided. The system will analyze the driving behaviors of surrounding vehicles surrounding an ADV, which is utilized to improve its original perception based on the analysis and to adapt to the updated driving environment. In one embodiment, in addition to the perception information provided by the sensors, the system analyzes the behaviors of the surrounding vehicles based on the perception information. Based on the behaviors of the surrounding vehicles, the system may detect there is may be an obstacle that has not been detected based on the perception information. Alternatively, the system may detect that an obstacle determined based on the perception information actually may not exist based on the behaviors of the surrounding vehicles. The paths created for the ADV may then adjusted accordingly to improve the autonomous driving of the ADV.