Driver Assistance Target Selection Using Lane and Trajectory Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driver assistance systems face challenges in accurately selecting target objects, particularly due to errors and ambiguities in back-projecting camera data into three-dimensional space, leading to incorrect target object selection and secondary lane interference issues, such as reacting to irrelevant objects like motorcyclists or parked vehicles.
Innovation Solution
A method for selecting a target object for a driver assistance system that considers the relevance of a foreign object in relation to both the ego vehicle's lane and its predicted trajectory, using image data from sensors like cameras or lidar, to adapt to the driver's intention and ensure collision-free behavior, without requiring explicit lane changes or relying on external features like direction indicators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If target object selection is based on back-projection of camera data into three-dimensional space, then target object detection capability is improved, but measurement precision deteriorates due to errors and ambiguities in back-projection
Solution Approach 1:
The patent processes camera data in two-dimensional image space rather than back-projecting into three-dimensional space. By maintaining operations in the native 2D measurement space of the image sensor, the system avoids back-projection errors while still achieving accurate target object selection through 2D relevance assessments.
2Reliability
If driver assistance system reacts to all detected objects in the lane, then detection completeness is improved, but reliability deteriorates due to secondary lane interference from irrelevant objects
Solution Approach 1:
The patent introduces the concept of relevance assessment that differentiates between relevant and irrelevant objects based on their spatial relationship to the ego vehicle's lane and predicted trajectory. Objects are evaluated locally according to their specific position and context, allowing the system to focus on relevant targets while ignoring irrelevant ones like parked vehicles or motorcyclists on lane boundaries.
Solution Approach 2:
The system changes the parameter of object evaluation from simple detection to relevance assessment based on spatial parameters. By introducing relevance criteria that consider the object's position relative to the lane and predicted trajectory, the system transforms the selection process to prioritize relevant objects while filtering out irrelevant detections.
3Reliability
If explicit lane changes or direction indicators are required to detect driver intention, then intention detection accuracy is improved, but response time deteriorates
Solution Approach 1:
The patent calculates a predicted trajectory of the ego vehicle in advance and uses it to assess object relevance before explicit driver actions occur. By evaluating objects against the predicted trajectory proactively, the system can detect driver intention earlier, without waiting for explicit lane changes or direction indicators, thus reducing response time while maintaining accuracy.
Data Source
AI summary
A method for selecting a target object for performing a function of a driver assistance system of an ego vehicle taking into account the target object. Image data of the surroundings of the ego vehicle generated by means of an image sensor are read in. In the method, a selection is made of a foreign object, detected by means of the image data, as a target object, taking into account at least: a first relevance of the foreign object in relation to a lane of the ego vehicle, and a second relevance of the foreign object in relation to a predicted trajectory of the ego vehicle. A device that is configured to carry out the corresponding method, and also to a corresponding computer program are also described.