Autonomous Vehicle Fallback Trajectories for Primary Compute Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in safely controlling themselves when there is a failure in their primary computing systems, especially on highways or in heavy traffic situations where pulling over or stopping may not be the best course of action.
Innovation Solution
A system comprising multiple computing systems, where a third computing system with reduced capabilities acts as a fallback to generate and send trajectories to the vehicle, based on the type of road and available sensor functionality, to ensure safe operation even if the primary system fails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a primary computing system is used to generate trajectories for autonomous vehicles, then the vehicle can operate with advanced functionality and maneuverability, but the system becomes vulnerable to failures that may compromise safety
Solution Approach 1:
The computing system is divided into multiple independent computing systems, each capable of generating trajectories. The primary computing system handles advanced maneuvers while a secondary computing system provides backup functionality. This segmentation ensures that if one system fails, the other can take over, maintaining both versatility and reliability.
Solution Approach 2:
A secondary computing system is pre-configured with the capability to generate trajectories in case the primary system fails. This beforehand cushioning ensures that safety is maintained by having a ready backup that can immediately take over trajectory generation without compromising vehicle operation.
2Reliability
If a fallback system applies brakes as hard and as quickly as possible in emergencies, then safety is prioritized, but the vehicle may stop in unsafe locations such as highways or heavy traffic
Solution Approach 1:
The fallback system dynamically adapts its emergency response based on the current driving context. When the vehicle is on a highway or in heavy traffic, the system modifies the emergency maneuver to avoid stopping in the lane, instead selecting alternative safe actions. This dynamic adaptation resolves the contradiction between prioritizing safety and avoiding harmful stopping locations.
Solution Approach 2:
The fallback system applies different emergency response strategies depending on the local driving context. On highways, it prevents lane stopping, while in other contexts it may apply standard emergency braking. This localized quality adjustment ensures safety while avoiding the harmful effect of stopping in unsafe locations.
3Measurement precision
If the primary computing system differentiates between different types of road users, then the vehicle can make more accurate behavior predictions, but the system complexity increases
Solution Approach 1:
The sophisticated road user differentiation capability is extracted from the secondary computing system and maintained only in the primary computing system. The secondary system focuses on basic trajectory generation without this complex differentiation functionality, reducing its complexity while the primary system maintains high measurement precision for road user identification.
Data Source
AI summary
Aspects of the disclosure relate to controlling a vehicle in an autonomous driving mode. The system includes a plurality of sensors configured to generate sensor data. The system also includes a first computing system configured to generate trajectories using the sensor data and send the generated trajectories to a second computing system. The second computing system is configured to cause the vehicle to follow a receive trajectory. The system also includes a third computing system configured to, when there is a failure of the first computer system, generate and send trajectories to the second computing system based on whether a vehicle is located on a highway or a surface street.