Control-Target Vehicle Setting via Image and Radar Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle driving assistance systems face challenges in accurately differentiating between stationary vehicles and stationary objects, leading to ineffective driving assistance control, particularly when stationary objects like manholes or overhead signs are misidentified as vehicles.
Innovation Solution
A control-target vehicle setting apparatus and method that integrates image and reflection point detection signals to determine whether a target is a vehicle or not, using a combination of movement history and data fusion processes to accurately set a control-target vehicle for driving assistance control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a stationary object is set as a control-target vehicle, then driving assistance control is performed, but the differentiation between stationary vehicle and stationary object other than vehicle becomes difficult
Solution Approach 1:
The patent segments the detection process into two independent detection units: an image detection unit that captures visual information and a radar detection unit that detects reflection points. By dividing the detection function across multiple units with different detection mechanisms, the system can cross-validate results and accurately differentiate between stationary vehicles and other stationary objects, resolving the contradiction between reliable control-target setting and precise differentiation.
2Adaptability or versatility
If only movement history is used to identify vehicles, then moving vehicles can be detected, but stationary vehicles cannot be identified
Solution Approach 1:
The patent merges the results from image detection and radar detection into a unified vehicle determination process. The image detection unit identifies vehicles based on visual characteristics while the radar detection unit provides complementary reflection point data. By combining these independent detection results, the system can identify both moving and stationary vehicles with high accuracy, eliminating the limitation of movement-based identification alone.
Data Source
AI summary
A control-target vehicle setting apparatus sets a control-target vehicle that serves as a target for driving assistance control. The control-target vehicle setting apparatus includes a detection signal acquiring unit and a setting control unit. The detection signal acquiring unit acquires a first detection signal that indicates a target by an image and a second detection signal that indicates a target by a reflection point. The setting control unit sets an acquired forward target as the control-target vehicle when the forward target is associated with an integration history that indicates that the forward target has been determined to be a vehicle using the first detection signal and the second detection signal in an integrated manner, even when the forward target is not associated with a movement history that indicates that the forward target has been detected as being a moving object.