Roadside Cloud Resource Allocation for Autonomous Vehicle Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Connected and Automated Vehicles (CAVs) face challenges due to high power consumption for sensing and control, limiting driving range and resource availability, and existing systems rely heavily on onboard components, which are expensive and complex.
Innovation Solution
A Device Allocation System (DAS) manages and allocates resources between Intelligent Roadside Toolbox (IRT) devices and CAVs to optimize reliability, intelligence, efficiency, and resilience, using cloud computing for resource distribution and management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If CAVs use onboard sensing and control systems, then vehicle autonomy and safety are improved, but power consumption increases and driving range is limited
Solution Approach 1:
The system segments the autonomous driving functionality into onboard components (sensors, processors, actuators) and offboard components (cloud-based servers, communication modules). Critical sensing and control functions are distributed between the vehicle and infrastructure, allowing the vehicle to maintain autonomy while offloading computationally intensive tasks to reduce onboard power consumption.
Solution Approach 2:
A communication intermediary (wireless communication system) is introduced between the CAV and the infrastructure. This intermediary enables the vehicle to receive control instructions and sensor data from offboard systems without requiring all processing to occur onboard, thereby reducing the vehicle's power consumption while maintaining autonomous operation.
2Measurement precision
If CAVs implement comprehensive onboard sensing systems, then detection accuracy and safety are improved, but system cost and complexity increase
Solution Approach 1:
The sensing system is segmented into onboard sensors (for immediate local detection) and offboard sensors/infrastructure (for extended range and supplementary data). This segmentation allows the vehicle to achieve comprehensive environmental awareness without equipping it with all necessary sensing capabilities onboard, thereby reducing system complexity while maintaining detection accuracy.
Solution Approach 2:
The system merges onboard sensing capabilities with offboard infrastructure sensing capabilities. By combining data from multiple sources (vehicle-mounted sensors and infrastructure sensors), the system achieves superior measurement precision without requiring each individual vehicle to carry a complete, complex sensing suite.
3Adaptability or versatility
If CAVs rely on individual vehicle-based control, then vehicle independence is maintained, but overall system efficiency and resource utilization decrease
Solution Approach 1:
The offboard control system provides universal services to multiple vehicles simultaneously. A single infrastructure-based control system can manage and coordinate numerous CAVs, optimizing traffic flow and resource allocation across the entire fleet rather than each vehicle operating independently, thereby improving overall system efficiency while vehicles retain their operational independence.
Solution Approach 2:
The system implements feedback loops where the offboard control system receives data from multiple vehicles, processes this information centrally, and sends coordinated control instructions back to individual vehicles. This feedback mechanism enables the system to optimize overall efficiency while maintaining vehicle independence through centralized coordination.
Data Source
AI summary
Provided herein is a Device Allocation System (DAS) for distributed autonomous vehicle-cloud operations. The technology provides a simplified DAS design in which an intelligent roadside toolbox is provided as a cloud-based platform of an Intelligent Roadside Infrastructure System. Cloud resources are used to supplement a connected and automated vehicle (CAV) to maintain the automation level of the CAV during abnormal driving conditions. Specifically, cloud resources comprise one or more of: computational resources, system security and backup resources, sensing resources, transportation behavior prediction and management resources, planning and decision-making resources, and/or vehicle control resources and/or instructions. Accordingly, the distributed autonomous vehicle-cloud operations maintain or restore the automation level of the CAV during abnormal driving conditions such as extreme weather or complex road geometry.