Automated Driving Complexity Metric and Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Level 4 automated driving systems are limited to small, geo-fenced areas and low-speed operations due to computational complexity, and there is a need to predict when automated driving functions may become unavailable for driver takeover or to allocate computing resources effectively.
Innovation Solution
An automated driving system that computes a complexity metric of an upcoming region using sensor data, adjusts operational modes by reallocating computational resources, and switches to external computing systems when necessary, while generating alerts for driver intervention if the system cannot determine a trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If Level 4 automated driving systems operate in complex environments, then the system can handle more driving scenarios, but computational complexity increases making the system unavailable
Solution Approach 1:
The system performs preliminary complexity assessment of the upcoming region before making automated driving decisions. By evaluating environmental complexity metrics in advance, the system can proactively determine when to switch operational modes or request driver takeover, preventing computational overload before it occurs.
Solution Approach 2:
The system dynamically adjusts its operational mode based on real-time complexity assessment. When the upcoming region is assessed as low complexity, the system operates in fully-automated mode; when complexity increases, it transitions to semi-automated mode or requests driver takeover, allowing the system to adapt its computational requirements to environmental conditions.
2Measurement precision
If the computing system processes high complexity sensor data, then detection accuracy improves, but system response time decreases
Solution Approach 1:
The system performs preliminary assessment of environmental complexity using sensor data before committing to full trajectory determination. By evaluating complexity metrics in advance, the system can prepare appropriate processing levels and allocate computational resources more efficiently, reducing actual response time while maintaining detection accuracy.
Solution Approach 2:
The system applies partial processing based on complexity assessment. For low-complexity regions, simplified processing suffices; for high-complexity regions, full processing is applied. This selective approach ensures adequate detection accuracy is achieved without always incurring the full computational time cost.
3Extent of automation
If the vehicle operates in fully-automated mode, then driver intervention is minimized, but computing resource requirements increase
Solution Approach 1:
The system dynamically adjusts its automation level based on real-time complexity assessment of the upcoming region. When complexity is low, the system operates in fully-automated mode with high computing resource utilization. When complexity increases, it transitions to semi-automated mode or requests driver takeover, reducing computing resource requirements while maintaining safety.
Solution Approach 2:
The system performs preliminary complexity assessment to predict when computing resources will be insufficient for fully-automated operation. By identifying these conditions in advance, the system can proactively reduce automation level or prepare for driver takeover, avoiding last-minute resource conflicts and ensuring smooth transitions.
4Productivity
If the system maintains high computing resources for Level 4 operation, then automated driving capability is improved, but system cost increases
Solution Approach 1:
The system dynamically adjusts computing resource allocation based on environmental complexity assessment. Rather than maintaining maximum resources continuously, the system allocates high computing resources only when needed for complex environments, and reduces allocation for simple environments, improving overall efficiency while maintaining Level 4 capability when required.
Solution Approach 2:
The system performs preliminary complexity assessment to predict upcoming computing resource needs. By identifying complex regions in advance, the system can pre-allocate computing resources efficiently, avoiding the need to maintain maximum resources continuously while ensuring adequate capacity is available when needed for Level 4 operation.
Data Source
AI summary
Technical solutions are described for controlling an automated driving system of a vehicle. An example method includes computing a complexity metric of an upcoming region along a route that the vehicle is traveling along. The method further includes, in response to the complexity metric being below a predetermined low-complexity threshold, determining a trajectory for the vehicle to travel in the upcoming region using a computing system of the vehicle. Further, the method includes in response to the complexity metric being above a predetermined high-complexity threshold, instructing an external computing system to determine the trajectory for the vehicle to travel in the upcoming region. If the trajectory cannot be determined by the external computing system a minimal risk condition maneuver of the vehicle is performed.