Vehicle Route Allocation Using Edge Computing for Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems (ADAS) and autonomous driving (AD) systems face challenges in processing large amounts of sensor data from multiple sensors in real time, which requires significant processing capacity, increasing vehicle costs and potentially reducing efficiency.
Innovation Solution
The system determines candidate routes and segments them into road segments, assessing the vehicle's computing resources and sensor configuration. It selects routes that allow external computing resources, such as vehicular edge computing devices (VECs), to offload computational tasks, ensuring the vehicle meets the computational thresholds for autonomous navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If multiple different types of sensors are included to enable semi-autonomous or fully autonomous driving, then the level of automated driving is improved, but the cost of the vehicle is substantially increased
Solution Approach 1:
The patent segments the computational workload by dividing routes into road segments and assessing computing resource requirements for each segment. This allows the system to allocate sensor and processing resources dynamically based on specific navigation needs rather than requiring all sensors to be active simultaneously, reducing overall system cost while maintaining autonomous driving capability.
Solution Approach 2:
The system performs preliminary assessment of computing resource requirements and sensor configuration needs before the vehicle reaches each road segment. By determining candidate routes and evaluating resource sufficiency in advance, the system can plan resource allocation ahead of time, allowing for cost-effective sensor selection while ensuring autonomous driving capability when needed.
2Reliability
If large amounts of sensor data from multiple different types of sensors are processed in real time to recognize obstacles and perform navigation functions, then the safety and navigation capability are improved, but the processing capacity requirement is substantially increased
Solution Approach 1:
The patent implements dynamic resource allocation where the system assesses the vehicle's computing resources and sensor configuration against the requirements of each road segment in real time. This dynamic assessment allows the system to adjust processing intensity and sensor activation based on actual navigation needs, reducing overall processing capacity requirements while maintaining safety.
Solution Approach 2:
The system changes operational parameters by evaluating different candidate routes and selecting routes where the vehicle's existing sensor and computing configuration can meet autonomous navigation thresholds. This parameter-based approach allows the system to maintain safety by selecting appropriate routes rather than requiring maximum processing capacity for all scenarios.
3Ease of manufacture
If the vehicle is equipped with limited number of sensors and processors to keep costs down, then the vehicle cost is reduced, but the ability to process sensor data in real time is insufficient for autonomous driving
Solution Approach 1:
The patent makes the computing resources universal by assessing their sufficiency across multiple candidate routes and road segments. The same limited processors are evaluated for their ability to handle different navigation scenarios, allowing the system to maximize the utility of limited computing resources across various driving conditions rather than requiring dedicated processing power for each scenario.
Solution Approach 2:
The system performs self-assessment of its computing resources and sensor configuration against autonomous driving requirements. By evaluating whether existing resources meet thresholds for specific road segments, the system can autonomously determine suitable routes without requiring additional hardware, allowing limited resources to serve multiple purposes effectively.
Data Source
AI summary
In some examples, a system may determine a plurality of candidate routes between a source location and a destination location for a vehicle, and may segment each candidate route into multiple road segments. Further, the system may receive vehicle computing resource information and sensor configuration information for the vehicle. The system may determine that at least one of the sensor configuration or the computing resources on board the vehicle fails to satisfy a threshold associated with autonomous navigation of a road segment of a first candidate route of the plurality of candidate routes. The system may select, for the vehicle, the first candidate route based at least on determining that a computing device external to the vehicle is scheduled to perform at least one computational task for the vehicle to enable the vehicle to meet the threshold associated with autonomously navigating the road segment of the first candidate route.