Autonomous Vehicle Redundant Computing Units for Limp Mode Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous vehicles, ensuring safety when a component of the automatic control system fails is crucial to prevent accidents, as existing systems lack effective redundancy and fail-safe mechanisms to maintain driving security.
Innovation Solution
A system comprising a master computing unit, a slave computing unit, lidars, cameras, millimeter wave radars, and a switch that provides redundant control capabilities, allowing the autonomous vehicle to enter a limp mode if either unit fails, ensuring safety by enabling the slave unit to take control based on environmental information from various sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single master computing unit is used to control the autonomous vehicle, then the device complexity is reduced, but the reliability deteriorates when the master computing unit fails
Solution Approach 1:
The patent implements a slave computing unit as a copy of the master computing unit's functionality. The slave unit contains the same computing modules (sensor processing module, recognition module, prediction module, planning module, and control module) and can independently execute the same control tasks. This copying approach ensures that if the master computing unit fails, the slave unit can take over immediately, thereby improving system reliability without significantly increasing complexity.
Solution Approach 2:
The system performs preliminary action by pre-configuring the slave computing unit with all necessary computing modules and capabilities before any failure occurs. The slave unit is kept in a standby state with identical functional architecture as the master unit, so that upon master unit failure, the slave unit can immediately activate and take control without requiring any reconfiguration or additional setup time, thus ensuring continuous safe operation.
2Reliability
If redundant computing units are added to ensure safety, then the reliability improves, but the device complexity increases
Solution Approach 1:
The control system is segmented into functionally independent but architecturally identical master and slave computing units. Each unit contains complete sets of computing modules (sensor processing, recognition, prediction, planning, and control modules) that can operate independently. This segmentation allows the system to achieve redundancy through modular duplication, where each module within both units can function autonomously, improving reliability while maintaining manageable complexity through clear functional separation.
3Reliability
If the switch provides fail-safe switching between master and slave computing units, then the reliability improves, but the device complexity increases
Solution Approach 1:
The switch acts as an intermediary component that mediates between the master computing unit, slave computing unit, and sensor systems. It provides intelligent fail-safe switching by monitoring the operational status of both computing units and automatically routing sensor data and control signals to the appropriate active unit. This intermediary approach simplifies the overall switching mechanism by centralizing the fail-safe logic in a dedicated component rather than distributing complexity across multiple systems.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Embodiments of the present disclosure relate to an autonomous vehicle and a system for the autonomous vehicle. The system includes a master computing unit (102) configured to control operations of the autonomous vehicle; a slave computing unit (104) communicatively coupled to the master computing unit (102) and configured to control the operations of the autonomous vehicle in response to detecting a failure of the master computing unit (102); at least one lidar (116, 118) configured to acquire environmental information around the autonomous vehicle; and a switch (106) communicatively coupled to the at least one lidar (116, 118), the master computing unit (102) and the slave computing unit (104), and configured to provide the environmental information to the master computing unit (102) and the slave computing unit (104) for controlling the autonomous vehicle. Since the master computing unit and the slave computing unit are provided, the autonomous vehicle 10 can enter the limp mode when any one of the master computing unit or the slave computing unit fails, to ensure the safety of the autonomous vehicle. In this way, at least L4 autonomous driving can be achieved.