Driving Mode Switch Verification Using Road Class and Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for switching between driver-controlled and autonomous driving modes lack sufficient precision and safety, particularly due to the inaccuracies of conventional satellite navigation, which can lead to safety-critical situations.
Innovation Solution
A method that uses satellite navigation to determine vehicle position and compare it with a road map to identify the road class, supplemented by environmental sensors to confirm the road class, ensuring accurate switching by evaluating weighted indications from various sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional satellite navigation is used for localization, then the system is simple and cost-effective, but the positioning precision is insufficient for safety-critical mode switching
Solution Approach 1:
The patent combines satellite navigation with sensor-based environmental recognition systems to achieve high-precision localization. The control device integrates data from both the satellite navigation system and environmental sensors (cameras, LIDAR, radar) to determine vehicle position and road class, compensating for satellite navigation inaccuracies through environmental feature matching.
Solution Approach 2:
The patent introduces environmental sensors as an intermediary layer between the satellite navigation system and the mode switching decision. The sensors capture environmental features (lane markings, traffic signs, road geometry) that serve as intermediate verification to confirm whether the vehicle is on a suitable road for autonomous driving, thereby improving positioning reliability without directly upgrading the satellite navigation hardware.
2Reliability
If environmental sensors are used to verify road class, then the reliability of mode switching increases, but the device complexity and cost increase
Solution Approach 1:
The patent makes the environmental sensors serve multiple functions: they are used both for road class verification and for general autonomous driving perception. The same sensor suite (cameras, LIDAR, radar) that is necessary for safe autonomous operation is also utilized to verify positioning accuracy and road suitability, avoiding additional dedicated hardware for mode switching verification.
Solution Approach 2:
The system implements feedback by continuously comparing satellite navigation position data with environmental features detected by sensors. The control device uses this feedback to confirm or reject the current road class classification, creating a closed-loop verification system that improves reliability while using existing sensor infrastructure.
3Reliability
If a specified area around the ascertained position is checked on the road map, then serious misinformation is compensated, but the processing time and computational load increase
Solution Approach 1:
The patent applies local quality by focusing the verification process on a specified area around the ascertained position rather than analyzing the entire road map. The control device extracts and compares environmental features only from the relevant local region, reducing computational load while maintaining verification accuracy for the critical mode switching decision.
Data Source
AI summary
A method for operating a vehicle, wherein a check is carried out to determine whether a switch of the driving mode can be safely carried out and a corresponding output is provided, wherein an estimation is made as to whether the road class traversed by the vehicle corresponds to a target road class, including the steps of: ascertaining the position of the vehicle using a satellite navigation system, comparing the ascertained position with a road map, wherein the road map includes an allocation of roads according to road classes, and determining whether a road of the target road class is located in a specified area around the ascertained position on the road map.