Vehicle Distributed Computing Architecture for Low-Latency Sensor Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in timely detection and processing of environmental data to make safe navigation decisions due to limitations in current distributed computing architectures, which can lead to delays and inefficiencies in sensor data processing and communication.
Innovation Solution
A scalable configurable chip architecture with a master embedded system and multiple slave embedded systems, each equipped with sensors, processes data in a distributed manner using high-speed interfaces, allowing for efficient data generation and processing without the need for synchronization through traditional protocols like IEEE 1588/gPTP, and enabling low latency communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional distributed computing architectures are used for sensor data processing, then system coverage and sensing capability are improved, but data processing delays and communication latency increase
Solution Approach 1:
The system divides sensor data processing into multiple independent processing units, each handling specific sensor types or data streams. This segmentation allows parallel processing of different sensor data simultaneously, reducing overall processing delay while maintaining comprehensive sensing coverage through the distributed architecture.
2Device complexity
If centralized processing architecture is used, then data processing is simplified, but communication bandwidth requirements and power consumption increase
Solution Approach 1:
The system implements local processing capabilities at distributed processing units, where each unit performs initial data processing and filtering locally before transmitting results to other units. This reduces the volume of data requiring communication across the network, thereby lowering power consumption and communication bandwidth requirements while maintaining processing effectiveness.
3Productivity
If high-speed communication protocols are implemented, then data processing efficiency is improved, but system costs and complexity increase
Solution Approach 1:
The system employs dynamic communication strategies where processing units adapt their communication frequency and data transmission rates based on real-time processing loads and data urgency. This dynamic approach maintains high data processing efficiency by prioritizing critical data transmission while reducing communication overhead during low-load periods, thereby avoiding the need for consistently high-speed protocols.
Data Source
AI summary
Provided are systems, methods, and computer program products for a distributed computing system for a vehicle. A vehicle (such as an autonomous vehicle) can have a hardware architecture including a master embedded system and multiple slave embedded systems. Each of the master and slave embedded systems can include, for example, a system on a chip (SoC). Each of the slave embedded systems can have multiple sensors of the vehicle assigned thereto, can process data generated by its assigned sensors, and can communicate an output of its processing to the master embedded system. The master embedded system can control timing of the slave embedded systems, such as by rotating sequentially through each of the slave embedded systems, such that timing of the sensors' data generating and processing can be controlled. Each of the slave embedded systems can communicate with the master embedded system via a high speed interface.