Autonomous Driving Sensor Zoning for Real-Time Drivable Area Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems face challenges in determining and monitoring their capability boundaries and associated risks in real-time, particularly in ensuring safety redundancy and minimum risk conditions, which is crucial for safe deployment of autonomous vehicles.
Innovation Solution
A computer-implemented method that calculates a zone failure risk score for each sensor zone, determines sensor capability coverage, and adjusts the drivable area based on obstacle position and predicted trajectories, using sensor data and map data to plan a safe trajectory for autonomous vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant and diversified sensors, hardware and algorithms are used to improve performance, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent segments the autonomous driving system into multiple independent sensor zones (e.g., front, rear, left, right zones) with dedicated sensors for each zone. This segmentation allows the system to maintain reliability through redundancy while managing complexity by organizing sensors into modular, manageable units rather than a monolithic system.
Solution Approach 2:
The patent introduces a spatial dimension to sensor deployment by defining specific zones and positioning sensors in three-dimensional space around the vehicle. This dimensional approach allows the system to achieve comprehensive coverage and redundancy without linearly increasing complexity, as sensors are strategically placed to maximize coverage efficiency.
2Reliability
If real-time monitoring of capability boundary and risk distribution is implemented, then safety is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary actions by pre-defining operational design domains (ODDs) and capability boundaries before real-time operation. Sensor zones and their coverage areas are pre-established, allowing the system to quickly assess whether current conditions fall within acceptable parameters without performing complex real-time analysis of every possible scenario.
Solution Approach 2:
The patent implements partial monitoring by focusing computational resources on critical zones and parameters rather than analyzing every aspect of the environment in full detail. The system monitors sensor functionality and coverage boundaries selectively, concentrating computational effort on areas that most impact safety while reducing monitoring intensity in less critical areas.
3Reliability
If sensor coverage boundary is strictly enforced to ensure safety, then reliability is improved, but area of stationary object decreases
Solution Approach 1:
The patent implements dynamic adjustment of drivable area boundaries based on real-time sensor functionality assessments. When sensors are functioning normally, the drivable area can be maximized. When sensor degradation or failure is detected in specific zones, the drivable area is dynamically reduced or adjusted to compensate for the reduced capability, maintaining safety while optimizing usable space.
Solution Approach 2:
The patent applies local quality adjustments by modifying drivable area constraints in specific zones rather than uniformly reducing the entire drivable area. If a sensor in a particular zone becomes degraded, only the drivable area in that specific zone is adjusted, while other zones maintain their full drivable area, thus preserving overall vehicle mobility while ensuring local safety.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Method for real-time monitoring of a safety redundancy autonomous driving system operating within a predefined risk tolerable boundary includes calculating a zone failure risk score for each of predetermined zones based on a sensor failure risk score associated with each of sensors mounted on the ADV. The predetermined zones being defined based on a sensor layout of the sensors. A sensor capability coverage of the ADV is determined based on the zone failure risk score associated with each of the predetermined zones. A drivable area of the ADV is determined based on the sensor capability coverage in view of map data associated with a current location of the ADV. A trajectory is planned based on the drivable area to autonomously drive the ADV to navigate a driving environment surrounding the ADV.