Autonomous Driving Route Retraction Under Remote Assistance Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote assistance systems for autonomous driving vehicles face challenges in maintaining accurate control when abnormalities occur, such as communication disruptions or performance degradation, leading to reduced reliability and safety.
Innovation Solution
The system sets a target retracting position along the vehicle's route, allowing the vehicle to safely stop with a margin, reducing the need for immediate emergency stops and minimizing the likelihood of requiring remote assistance, by selecting a position within a retracting margin section that includes specific positions where assistance may be needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous driving vehicle immediately stops when an abnormality occurs in the remote assistance system, then the safety of the vehicle is improved, but the continuity of autonomous driving is reduced and the vehicle may stop at unsafe locations
Solution Approach 1:
The system pre-identifies safe stopping positions (specific positions) along the target route before an abnormality occurs. When an abnormality is detected, the vehicle can immediately navigate to the nearest pre-identified safe position rather than stopping at the current location, which may be unsafe. This preliminary preparation of stopping positions resolves the contradiction by ensuring safety while maintaining continuity of the journey.
Solution Approach 2:
The system introduces an intermediary safe stopping position between the current vehicle location and the final destination. Instead of directly stopping at the destination or the current location, the vehicle stops at an intermediate safe position that has been pre-identified. This intermediary position acts as a buffer that ensures safety while allowing the journey to continue afterward, thus maintaining continuity.
2Productivity
If the autonomous driving vehicle travels to the destination before stopping, then the productivity is improved, but the safety is reduced when an abnormality occurs in the remote assistance system
Solution Approach 1:
The system pre-identifies safe stopping positions along the route before the vehicle reaches them. This allows the vehicle to maintain high productivity by continuing to the destination under normal conditions, while having safe stopping positions ready in advance if an abnormality occurs. The preliminary identification of stopping positions does not interfere with normal efficient operation but provides a safety net when needed.
Solution Approach 2:
The system dynamically adjusts the stopping behavior based on the operational status of the remote assistance system. When the system is normal, the vehicle continues to the destination without unnecessary stops, maintaining productivity. When an abnormality is detected, the vehicle dynamically changes course to stop at the nearest pre-identified safe position, ensuring safety. This dynamic adjustment resolves the contradiction between productivity and safety.
3Reliability
If the autonomous driving vehicle stops at the current position when an abnormality occurs, then the safety is improved, but the device complexity increases due to immediate emergency stop requirements
Solution Approach 1:
The system pre-identifies and stores safe stopping positions along the target route before an abnormality occurs. This preliminary action simplifies the control logic during an abnormality event - the vehicle only needs to navigate to the nearest pre-identified position rather than performing complex real-time analysis to determine where to stop. This reduces the computational complexity and control system burden during critical moments.
Solution Approach 2:
The system uses the vehicle's own navigation and path planning capabilities to identify and navigate to safe stopping positions along its intended route. Rather than requiring an external emergency stop system, the vehicle leverages its existing autonomous driving functions to safely stop at pre-identified positions, reducing the need for additional complex emergency stop hardware and control logic.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An autonomous driving system (10) includes one or more storage devices (120) configured to store specific position information (220) indicating specific positions (PS) at which there is a possibility that remote assistance is required; and one or more processors (110) configured to: determine presence or absence of an abnormality in a remote assistance system (4) configured to provide the remote assistance to an autonomous driving vehicle (1), set, when an abnormality is detected in the remote assistance system (4), any one of the specific positions (PS) on a target route (RT) from a current position (P1) of the autonomous driving vehicle (1) to a destination (DST) as a limit position (PL) based on the specific position information (220); set a target retracting position (PE) such that the target retracting position (PE) is included in the target route (RT) from the current position (P1) to the limit position (PL); and control the autonomous driving vehicle (1) such that the autonomous driving vehicle (1) stops at the target retracting position (PE).