Lane centering AI for Autonomous Vehicles
Autonomous driving nowadays become the reality not a dream and a driverless vehicle or an autonomous vehicle is the one that is capable to operate itself and perform all the necessary and required functions without any human intervention, through ability to sense its surrounding. for this lane cente
2025-06-28 16:33:57 - Adil Khan
Lane centering AI for Autonomous Vehicles
Project Area of Specialization Artificial IntelligenceProject SummaryAutonomous driving nowadays become the reality not a dream and a driverless vehicle or an autonomous vehicle is the one that is capable to operate itself and perform all the necessary and required functions without any human intervention, through ability to sense its surrounding. for this lane centering is a vast domain in it. The lane centering is a mechanism designed to keep a car centered in the lane, relieving the driver of the task of steering. The first commercially available lane centering systems were based on the “off the shelf” systems which is created by Nissan propilot and tesla Autopilot.
The lane detection system used by the lane departure warning system uses image processing techniques to detect lane lines from real-time camera images fed from cameras mounted on the automobile. Examples of image processing techniques used include the Hough transform, Canny edge detector, Gabor filter and deep learning. we are also use the concepts of deep learning to implement this lane centering after detecting the other lanes and then choice our egolane. The invention of this systems births a lot of benefits like Even just a fleeting moment of inattention can lead a driver to depart from their lane. The lane centering assist prevents this by actively keeping the vehicle in the center of its lane, bringing both driver and load to their destination safely and comfortably and it is also beneficial in case of business as the cost of labor is also saved if this autonomous vehicle is implemented.
Our work and studies typically focused on improving the control accuracy of the autonomous driving vehicles. In addition to the control accuracy, driver/passenger comfort is also an important performance measure of the system. The Goal of our project is to Ensure the safe operation and functional safety of reliable automated lane centering control systems and some objectives of our project are:
• Provide research findings supportive of improving driver awareness and training.
• Provide research findings supportive of functional safety concepts and requirements, including diagnostic needs, identify performance parameters, functional safety test scenarios o driver-vehicle interface requirements.
Project ObjectivesThe objective of this project was to design and develop a lane detection algorithm for autonomous vehicles applications. This system is an essential part of the advanced driver assistance systems (ADAS) used in autonomous vehicles. This feature is responsible for detecting lanes, measuring curve radius. The self driving car market is growing at a very fast pace. Many companies are working in this problem trying to solve every aspect of it, so that autonomous cars can drive safely on the roads. It is a very complex problem due to the many aspects that it relies on: robotics, path planning, navigation, computer vision, mechanics, etc. This project is focused in the computer vision aspect of it, a crucial module. If an automated car is going to drive around unpredictable environments, it has to be able to perceive and detect every small detail that surrounds it. So for this project, a lane detection algorithm is proposed as part of the perception component for a self driving vehicle. By using a video feed input of a car driving on the highway, the algorithm will detect where the lane is so that the car can use its location to avoid getting out of the it. It will also be able to predict any turn on the road to secure a good tracking of the lane.
This work is based on computer vision to develop an autonomous or driverless vehicle,
we need to determine the algorithms which will efficiently work and assist using artificial
intelligence, machine learning, image processing and computer vision modules.
Algorithms can be used to:
1. Detect Lanes: To detect the position and parameters by image processing to
detect lane and region
2. Drive-able Region: A region where a car can move safely
3. Lane Centering (LC): Automatically steers the vehicle to maintain a central
path within the lane.
4. Lane Departure Warning (LDW): Alerts the driver to an unindicated (and
therefore presumably unintended) lane departure.
5. Lane Keeping Assist (LKA): Automatically steers the vehicle to stay within
lane boundaries.
An important component of such a system is the evaluation of image sequences recorded with cameras mounted in a moving vehicle. These image sequences provide information about the vehicle’s environment which has to be analyzed for drivable region. An image processing technique called Sobel edge detection is applied to each image for detecting boundary lines in the image plane. To get the angle of inclination of lane markings a Standard Hough Transform
technique is applied to images. Then their ratio of left lane markings to the right lane markings gives the vehicle’s position with respect to the center line of the road
Using image processing we will convert our colorful image into grayscale, after applying canny algorithm, we will use Gaussian filter to smooth the image, which will reduce the noise and convert three channel images into single channel, which will result in less computation and processing. In the end we will use Hough transformation to detect the lines.
We will design a lane assistant which will make sure our vehicle is moving in the desirable/drivable region. If it goes against that it will notify the system to take over the control.
For the line detection to be effective the camera must be supplied with the following quantities describing the motion of the vehicle:
? lateral acceleration
? longitudinal acceleration
? speed
? steering wheel angle
? wheel velocity
For detecting lanes continuously, the speed of vehicle should be higher so that the boundary lane will be captured in most frames. There are usually several times during the driving when the driver crosses the lane markings intentionally: to cut a curve, to overtake another vehicle, or to let other vehicle pass for example. It may be a part of the behavior to cross the lane markings often, which causes a lot of false warnings. The derived lane related parameters can be used for controlling the steering angle of the vehicle. But here system would obviously assist the car, decision to move or steer is totally at AI module of steering.
? The purpose of the camera is assisting the driver which means the following functions:
Lane keeping support
Lane departure warning
? For Motion Perception at the core of Advanced driver-assistance systems (ADAS) are camera sensors integrated into a vehicle that constantly scan the road and stream footage. Using real-time inference, the onboard detects road features and driving hazards and if necessary, alerts the driving system, or, in the case of more advanced active systems, directs the vehicle to actively react to prevent
unsafe driving and collisions.
Customer’s safety has always been central area of interest. Research has shown that more
than 94% of road accidents or crashes occur due to risky and dangerous behaviors of
driver. So, autonomous vehicles are designed in such a way that they can provide an
additional safety to the passengers by intelligent AI design, which will result in fewer bad
incidents. A fully automated vehicle is designed to make roads safe and law ruling, even
disabled person, like blind, can travel and walk around easily. These vehicles can save
our time in many ways, like there will be fewer traffic jams, lesser rules and signal
violation so eventually travelling will consume lesser amount of time and fuel
consumption can also be reduced somehow, finally which can create lesser pollution too.
So, concluding this we have all kinds of human and environmental benefits from this type
of vehicle.
It helps keep passengers safer on the roads, reduces the risks of traffic accidents, saves lives and has the potential to revolutionize the driving experience by enabling autonomous driving.
Beneficiaries of the Project
? Personal use vehicles
? Online Cab Booking Services
Goods Transferring vehicles
? Public transportation
The project will provide following outputs:
1. Lane Keeping Assistant
2. Lane Departure Warning
3. Lane Centering
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 20000 | |||
| Cameras | Equipment | 1 | 10000 | 10000 |
| Usage of car | Miscellaneous | 1 | 5000 | 5000 |
| Printing | Miscellaneous | 1 | 5000 | 5000 |