Lane Detection Algorithm in Curves Based on Multi-Sensor Fusion

Area of Science:

• Computer Vision
• Robotics
• Automotive Engineering

Background:

• Lane detection is crucial for advanced driver-assistance systems (ADAS) and autonomous driving.
• Traditional sliding window algorithms struggle with high-curvature road sections.
• Large curvature curves are common and challenging scenarios for current lane detection systems.

Purpose of the Study:

• To enhance the performance of sliding window lane detection algorithms in large curvature curves.
• To develop a more robust lane detection method for assisted and autonomous driving systems.
• To improve the tracking and recognition of lane markings in challenging road geometries.

Main Methods:

• An improved sliding window lane detection algorithm is proposed.
• Integration of steering wheel angle sensors and binocular cameras.
• Utilizing previous frame's steering wheel angle to predict the search center of sliding windows.
• Employing a binocular camera to assist in initial sliding window positioning.

Main Results:

• The improved algorithm demonstrates superior performance in recognizing and tracking lane lines on high-curvature bends.
• Simulation and experimental results validate the effectiveness of the proposed method.
• The integration of sensor data enhances lane detection accuracy compared to traditional methods.

Conclusions:

• The proposed sliding window lane detection method effectively addresses the limitations of traditional algorithms in large curvature curves.
• The integration of steering wheel angle and binocular camera data significantly improves lane tracking robustness.
• This advancement contributes to safer and more reliable autonomous driving systems.