Remote Takeover Control for Autonomous Vehicle Critical Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face limitations in detecting and managing critical events on their route, leading to potential errors in piloting actions due to rapid analysis and limited possibilities for sudden changes in direction or braking, especially in unforeseen situations like accidents or adverse weather conditions.
Innovation Solution
A method that involves predictive detection of critical events by a remote control center using combined data from traffic lane conditions, weather, and vehicle sensors, allowing an operator to take temporary remote control of the vehicle to implement optimal steering actions, with a risk assessment algorithm and data weighting to prioritize critical conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the autonomous driving system performs rapid analysis of the environment using sensors and AI algorithms, then the vehicle can detect and respond to unforeseen situations quickly, but this rapid analysis can lead to misinterpretations and inappropriate driver actions
Solution Approach 1:
The system performs preliminary detection and classification of events upstream before the vehicle reaches them. By analyzing traffic lane conditions, weather data, and sensor information in advance, the system prepares appropriate piloting actions beforehand, allowing slower but more accurate human-like decision-making rather than forced rapid responses
Solution Approach 2:
The patent introduces an intermediary layer between sensor detection and autonomous response by comparing sensor data with pre-acquired reference data from maps and databases. This intermediary verification step helps prevent misinterpretations by cross-checking detected events against known environmental information before triggering autonomous piloting actions
2Extent of automation
If the autonomous driving system relies solely on onboard sensors and rapid AI analysis, then the system maintains full automation without human intervention, but the system has limited possibilities for sudden changes in direction or braking when facing critical events
Solution Approach 1:
The system introduces a remote control center as an intermediary between the autonomous vehicle and human operators. When critical events are detected upstream, the system can transfer control to a human operator who provides flexible, adaptive piloting actions that complement the autonomous system's capabilities
Solution Approach 2:
The system dynamically adjusts the level of automation based on the situation. For routine operations, full automation is maintained. When upstream detection identifies critical events requiring complex judgment or sudden maneuvers, the system dynamically transitions to human-operated mode, creating a flexible hybrid control architecture
3Loss of time
If the system detects critical events only in the immediate vicinity of the vehicle, then the response time is minimized, but the possibilities for driver intervention are limited and misinterpretations occur more frequently
Solution Approach 1:
The system performs preliminary detection of critical events upstream along the vehicle's route using traffic lane condition data, weather information, and sensor data from other vehicles. By detecting events before the vehicle reaches them, the system gains both early warning time and the opportunity to verify event accuracy through multiple data sources before triggering responses
4Productivity
If the autonomous driving system makes independent driving decisions in real-world traffic conditions, then the vehicle can operate autonomously without driver intervention, but errors in interpreting critical events can lead to inappropriate piloting actions
Solution Approach 1:
The system incorporates feedback loops where detected events are cross-validated against reference data from maps and databases before triggering autonomous responses. This feedback mechanism allows the system to verify event authenticity and adjust its interpretation, reducing errors while maintaining autonomous operation
Solution Approach 2:
By performing preliminary analysis of traffic conditions, weather, and sensor data upstream, the system prepares accurate event interpretations in advance. This preliminary processing reduces the cognitive load during actual response moments and minimizes interpretation errors while maintaining efficient autonomous operation
Data Source
Figure 1
AI summary
The invention relates to a method for assisting an autonomous motor vehicle, comprising the following steps: - using an autonomous motor vehicle equipped with an automatic driving device which is adapted to decide driving actions to be carried out on said vehicle in order to autonomously circulate said vehicle on a route, - connecting the automatic driving device to a computer server of a remote control center, said connection being produced through a wireless communication network, - determining the geographical position of the vehicle by means of the computer server, characterized in that: - the method comprises a step of predictive detection of a critical event on the route upstream of the geographical position of the vehicle, which detection results in the handgrip of the driving device by an operator of the control center, so that said operator takes temporarily the control of the remote vehicle and decides driving actions, - predictive detection of the critical event resulting from the combined analysis of data by the computer server, which data comprises: - data on the state of the traffic lanes taken on the route upstream of the geographical position of the vehicle, and - data on the road conditions on the route upstream of the geographical position of the vehicle, and - data on the meteorological conditions on the route upstream of the geographical position of the vehicle, and - data coming from sensors installed on the vehicle.