Autonomous Vehicle Fallback Task Control for Safe Self-Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in safely operating when they cannot complete primary tasks due to malfunctions or low resources, lacking executable instructions, and there is a need for fallback tasks to ensure safe operation and maintenance without human intervention.
Innovation Solution
A system with processors and memory that manage fallback tasks by receiving status updates, determining trigger conditions, and executing fallback instructions such as driving to a base location or maintenance depot, allowing the vehicle to autonomously navigate to a safe location for servicing or reconnection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If autonomous vehicles operate without human drivers, then productivity and transportation efficiency are improved, but reliability deteriorates due to malfunctions and lack of human intervention
Solution Approach 1:
The system pre-configures fallback tasks and trigger conditions before autonomous operation begins. When malfunctions occur or resources are depleted, the vehicle automatically executes pre-planned fallback procedures without human intervention, ensuring continuous safe operation while maintaining high productivity
Solution Approach 2:
The autonomous vehicle monitors its own status, detects malfunctions and resource levels, and autonomously executes fallback tasks without human assistance. The system self-manages safety protocols, maintaining reliability while enabling unmanned productive operation
2Reliability
If fallback tasks are implemented for autonomous operation, then reliability is improved through safe fallback procedures, but device complexity increases due to additional monitoring and execution systems
Solution Approach 1:
The fallback task management system integrates multiple functions into a unified architecture: status monitoring, trigger condition evaluation, task selection, and execution control all operate within the existing autonomous vehicle control system, avoiding significant complexity increases while ensuring reliable safe operation
3Reliability
If continuous monitoring of status updates is performed, then reliability is improved through early detection of trigger conditions, but use of energy increases due to constant system monitoring
Solution Approach 1:
The system performs status monitoring and trigger condition evaluation at periodic intervals rather than continuously, reducing energy consumption while maintaining reliable detection of malfunctions and resource depletion conditions that require fallback task execution
4Ease of operation
If fallback tasks are executed autonomously without human intervention, then ease of operation is improved, but loss of time occurs when fallback tasks interrupt primary tasks
Solution Approach 1:
The system adjusts the priority and execution timing of fallback tasks based on the severity of trigger conditions and current vehicle state. Less critical fallback tasks are scheduled to execute after primary tasks complete, while critical safety-related fallback tasks interrupt primary tasks immediately, optimizing both autonomous operation and task completion time based on dynamic parameter changes
Data Source
AI summary
Aspects of the present disclosure relate to a system having a memory, a plurality of self-driving systems for controlling a vehicle, and one or more processors. The processors are configured to receive at least one fallback task in association with a request for a primary task and at least one trigger of each fallback task. Each trigger is a set of conditions that, when satisfied, indicate when a vehicle requires attention for proper operation. The processors are also configured to send instructions to the self-driving systems to execute the primary task and receive status updates from the self-driving systems. The processors are configured to determine that a set of conditions of a trigger is satisfied based on the status updates and send further instructions based on the associated fallback task to the self-driving systems.