Autonomous Vehicle Control Failover Using Dual Processing Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in maintaining control and fault tolerance due to potential software or hardware faults in their control systems, which can lead to loss of functionality and safety issues.
Innovation Solution
A control system interface with a primary and secondary processing unit, along with a fault detection module, is implemented to detect faults and seamlessly transition control from the primary to the secondary processing unit, ensuring continuous autonomous operation even in the presence of faults, by receiving and processing trajectories to generate control signals for steering, braking, and acceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single processing unit is used for autonomous vehicle control, then the device complexity is low, but the reliability is insufficient due to potential software or hardware faults
Solution Approach 1:
The control system is segmented into multiple independent processing units (primary and secondary), each capable of independently processing trajectories and generating control signals. This segmentation allows the system to maintain functionality even when one unit fails, directly resolving the contradiction by improving reliability through structural division.
Solution Approach 2:
Different processing units are assigned different functional roles (primary vs. secondary) with specialized capabilities. The primary processing unit handles normal operations while the secondary unit is prepared for fault conditions. This local differentiation of quality and function enables the system to achieve high reliability without requiring complete redundancy of all components.
2Reliability
If redundant processing units are implemented for fault tolerance, then the reliability improves, but the device complexity increases
Solution Approach 1:
The secondary processing unit is pre-configured and prepared in advance to take over control functions if the primary unit fails. This preliminary preparation includes having the secondary unit ready to process trajectories and generate control signals without requiring complex real-time reconfiguration, thus improving reliability while limiting the increase in device complexity.
Solution Approach 2:
A fault detection module acts as an intermediary between the primary and secondary processing units. This mediator monitors the primary unit's health and automatically triggers the failover process when faults are detected, eliminating the need for complex manual or software-based switching mechanisms and keeping the overall system complexity manageable.
3Reliability
If fault detection and switching mechanisms are added, then the reliability is improved, but the device complexity increases
Solution Approach 1:
The fault detection module continuously monitors the primary processing unit and provides feedback about its operational status. This feedback mechanism enables automatic detection of faults and triggers the appropriate response (switching to secondary unit) without requiring complex diagnostic algorithms or multiple monitoring layers, thus improving reliability while maintaining simplicity.
Data Source
AI summary
A control system for a vehicle includes a time sensitive network switch and one or more processors. The time sensitive network switch is configured to receive sensor data from light detection and ranging (LIDAR) sensors, radio detection and ranging (RADAR) sensors, or image sensors. The sensor data is representative of estimated objects that are in an external environment of the autonomous vehicle. The one or more processors are configured to receive the sensor data and determine instructions to control movement of the autonomous vehicle at least partially based on the sensor data from the time sensitive network switch and at least partially based on map data.