Autonomous Vehicle Lane Keeping via Deceleration Curb Angle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in accurately detecting lane departure, especially when lanes are not clearly marked, leading to potential deviations from the intended path.
Innovation Solution
A computer-implemented method that detects contact with a deceleration curb using sensors, calculates the angle between the vehicle's movement direction and the lane direction, and generates control commands to adjust the vehicle's direction to stay within the lane by determining the distance traveled and wheel positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional lane detection methods are used, then the system can detect lane markings, but it fails when lanes are not clearly marked or painted with insufficient contrast
Solution Approach 1:
The patent introduces deceleration curbs as intermediary physical markers across the lane. These curbs serve as tactile and positional references that the vehicle's sensors can detect through wheel contact. When the vehicle contacts the curb, the system calculates the angle of departure from the lane direction based on the contact points of left and right wheels, providing reliable lane departure detection independent of visual lane markings.
Solution Approach 2:
The patent replaces the optical/visual detection system (cameras and image processing for lane markings) with a mechanical detection system. The vehicle's wheel-curbs interaction provides direct physical feedback about lane position and departure angle, substituting the need for visual processing of lane markings with tactile and positional mechanical sensing.
2Extent of automation
If the vehicle operates autonomously without clear lane markings, then it can navigate independently, but it cannot accurately detect lane departure or maintain correct lane position
Solution Approach 1:
Deceleration curbs serve as intermediary reference objects that enable the autonomous vehicle to measure its lane position and departure angle. By detecting the contact points of left and right wheels with the curb and calculating the resulting angle, the system achieves precise lane position measurement without requiring visual lane markings, thus maintaining autonomous operation accuracy.
3Measurement precision
If rumble strips are used for lane detection, then the system can estimate vehicle heading, but it requires specific road infrastructure with rumble strips
Solution Approach 1:
The deceleration curb system serves multiple functions: it acts as a lane boundary marker, provides tactile feedback for wheel contact detection, and enables angle calculation for lane departure detection. This multi-functional approach replaces the specialized rumble strip system, making the solution applicable to broader road types without requiring pre-installed acoustic rumble strips.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A lane departure detection system (306) detects that an autonomous driving vehicle (ADV) is departing from the lane (500) in which the ADV is driving based on sensor data captured when the ADV contact a deceleration curb (502) such as a speed bump laid across the lane (500). When the ADV contacts the deceleration curb (502), the lane departure detection system (306) detects and calculates an angle (505) of a moving direction (504) of the ADV vs a longitudinal direction of the deceleration curb (502). Based on the angle (505), the system calculates how much the moving direction (504) of the ADV is off compared to a lane direction (503) of the lane. The lane direction (503) is typically substantially perpendicular to the longitudinal direction of the deceleration curb (502). A control command such as a speed control command and/or a steering control command is generated based on the angle (505) to correct the moving direction (504) of the ADV.