Automated Lateral Guidance at Road Branches Using Preceding Vehicle Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lateral guidance systems for vehicles are inadequate for hands-off mode due to insufficient quality of lane information from environment sensors, leading to reduced controllability and safety concerns, especially in conditions like low-lying sun or heavy rain.
Innovation Solution
A method that combines satellite-based position data with high-resolution map data to determine the vehicle's position relative to lanes, and uses sensor data from environment sensors, such as cameras and radar, to detect preceding vehicles and road branches, thereby improving lane recognition and issuing takeover requests when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If sensor data from environment sensors (camera, radar) is used for lane recognition, then the system can provide automated lateral guidance, but the quality of lane information deteriorates under environmental influences (low-lying sun, heavy rain)
Solution Approach 1:
The patent introduces map data as an intermediary information source that mediates between the unreliable sensor data and the lane recognition system. When sensor data quality deteriorates due to environmental conditions, the system switches to using pre-stored map data to maintain reliable lane recognition and automated guidance.
Solution Approach 2:
The system performs preliminary actions by pre-storing high-resolution map data including lane information, road geometry, and environmental features before they are needed. This allows the system to quickly switch to reliable map-based lane recognition when real-time sensor data becomes unreliable during adverse weather conditions.
2Ease of operation
If hands-off mode is implemented to relieve driver burden, then ease of operation improves, but controllability and safety deteriorate
Solution Approach 1:
The system implements continuous feedback by monitoring the quality and reliability of lane information from multiple sources (sensors and map data). When lane recognition confidence drops below a threshold, the system provides feedback to the driver through takeover requests, ensuring safety while maintaining hands-off operation under normal conditions.
Solution Approach 2:
The system prepares takeover requests in advance when lane information quality deteriorates, ensuring that controllability is restored before safety is compromised. This preliminary action allows the system to maintain hands-off mode as long as possible while having safety mechanisms ready to activate.
3Measurement precision
If map-based position determination is used to determine vehicle position with lane accuracy, then position precision improves, but errors and variances can still arise during operation
Solution Approach 1:
The patent merges map-based position determination with sensor-based lane recognition and preceding vehicle trajectory detection. By combining multiple independent position estimation methods, the system achieves both high precision and operational reliability, as errors in one method can be compensated by the others.
Solution Approach 2:
The system uses feedback from multiple position determination methods to validate and correct map-based position data. When sensor data or preceding vehicle trajectories indicate deviations from map-based predictions, the system adjusts its position estimation to maintain operational consistency and accuracy.
Data Source
AI summary
A method for assisting a user of a vehicle during an automated lateral guidance of the vehicle on a road having multiple lanes includes determining map-based position data that describes a present position of the vehicle in relation to the lanes on the basis of satellite-based position data and high-resolution map data, receiving sensor data from an environment sensor of the vehicle, detecting a preceding vehicle and a trajectory taken by the preceding vehicle based on the sensor data, determining probabilities of occupancy for the vehicle for at least some of the lanes based on the map-based position data and/or sensor data, recognizing a branch in the road based on the map-based position data and/or sensor data, and actuating an output device for outputting a takeover request before the branch is reached according to the probabilities of occupancy and the trajectory of the preceding vehicle.