Autonomous Driving Task Offloading Across Vehicle Edge Cloud Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack an optimal solution for offloading autonomous driving tasks across the vehicle, edge, and cloud computing layers, leading to inefficiencies in computing resources and increased costs.
Innovation Solution
A system and method that utilize a modeling module to create a system model incorporating communication latency between computing nodes, and an allocation module to allocate autonomous driving tasks across vehicle, edge, and cloud nodes, minimizing end-to-end latency while ensuring safety requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computational tasks are processed only at the vehicle end, then autonomous driving functionality is achieved, but vehicle cost increases significantly due to high-performance sensors and computing chips
Solution Approach 1:
The patent segments autonomous driving tasks into different computational categories and distributes them across vehicle-end, edge-end, and cloud-end computing nodes. Critical real-time tasks remain at the vehicle end, while less time-sensitive tasks are offloaded to edge and cloud nodes, reducing vehicle hardware requirements and cost.
Solution Approach 2:
The patent introduces edge computing nodes as intermediaries between vehicles and cloud infrastructure. These edge nodes handle intermediate computational tasks, reducing the burden on vehicle hardware while providing faster response than pure cloud processing for certain tasks.
2Device complexity
If tasks are offloaded to edge or cloud ends, then vehicle cost is reduced, but end-to-end latency increases
Solution Approach 1:
The patent applies local quality by assigning different task types to different computing locations based on their latency requirements. Time-critical perception and control tasks are processed locally at the vehicle end, while less time-sensitive mapping and planning tasks are offloaded to edge or cloud nodes, optimizing the overall system latency.
Solution Approach 2:
The patent implements dynamic task offloading decisions that adapt to real-time conditions. The system dynamically determines which tasks to offload based on current latency requirements, computational load, and network conditions, allowing flexible optimization between cost and latency.
3Device complexity
If task offloading is implemented without optimization, then vehicle hardware requirements are reduced, but computing resource utilization efficiency decreases
Solution Approach 1:
The patent incorporates feedback mechanisms where the system continuously monitors task completion times, resource utilization, and latency metrics. Based on this feedback, the task offloading strategy is dynamically adjusted to optimize computing resource utilization efficiency while maintaining reduced vehicle hardware requirements.
Solution Approach 2:
The patent changes key parameters such as task offloading thresholds, computational task categorization, and resource allocation strategies to optimize the balance between vehicle hardware requirements and computing resource utilization efficiency across the distributed system.
Data Source
AI summary
A system and method for offloading autonomous driving tasks is disclosed. The system includes a plurality of computing nodes comprising one or a plurality of computing nodes located on a vehicle, one or a plurality of computing nodes located on edge devices, and one or a plurality of computing nodes located on cloud devices. The system further includes a modeling module configured to create a system model that comprises a communication latency between each pair of computing nodes among the plurality of computing nodes. The system also includes an allocation module configured to allocate a plurality of autonomous driving tasks of an autonomous driving service based on the system model in order to offload each autonomous driving task to one of the plurality of computing nodes, wherein the allocation is performed such that the end-to-end latency of the autonomous driving service is minimized.