AI / Machine Learning

Challenges of the Automotive Industry

Quality control is a vital process in all industries to ensure that the product they deliver are of high quality and without defect. In the automotive industry this is of critical importance both from a production perspective as well as from customer safety concerns. Despite spending billions of dollars and utilizing third party quality checks, it is still a recurring problem in the automotive industry where expensive recalls are often made.

 

INTRODUCING AI SOFTWARE

Visual inspections of parts are vital to catch defects in the early stage as even the smallest defect could potentially be the root cause of accidents or expensive recalls. Investing in quality control systems is essential in the automotive industry to ensure reliability, boost production efficiencies and reduce cost. Abraham Innovations sets the standard for vision inspection with our newly released AI software. Our software is a powerful turnkey solution that delivers fast, reliable, and accurate results. The software optimizes the model through continuous training processes and analysis of images and datasets. 

APPLICATION  EXAMPLE  #1

READING OF VEHICLE IDENTIFICATION NUMBERS
(VIN) FROM CHASSIS IN FINAL ASSEMBLY LINE

Vehicle Identification Numbers (VIN) is a unique number that consists of a combination of 17 letters and digits. It provides key information about the vehicle’s manufacturer, model, model year, make, equipment, class, vehicle history etc. Recent changes in requirements for key foreign markets mean strict penalties can be imposed if VIN marks are not in compliance.
AI software uses the character recognition algorithm to verify that the VIN number is clear, complete, legible, and correct. Any suspect VIN numbers are flagged and verified by manual inspection.

APPLICATION  EXAMPLE  #2

AUTOMATED VEHICLE BODY INSPECTION IN
AUTOMOTIVE PAINT LINE

Majority of the quality inspections are performed by auditors using human eyes. Detecting surface defects on glossy surfaces such as paint, metal, glass, and others have always been a difficult task for industry. Waviness, scratches, discolored specks are all very hard to spot even by a skillful quality inspection auditor.
AI software is equipped with sophisticated image processing algorithms that are optimized for different types of surfaces. Coupled with a multiplecamera, PC-based machine vision system with special light scheme, all the aforementioned surface defects become detectable. The software delivers high detection rates and reliable results in 24/7 mode and supports all types of colors and surfaces.

APPLICATION  EXAMPLE  #3

AUTOMATED VEHICLE BODY INSPECTION IN
AUTOMOTIVE BODY-IN-WHITE (BIW) ASSEMBLY LINE

Body-in-white (BIW) refers to a stage in the automotive industry in which a car body’s sheet metal components have been welded and assembled together to form the vehicle’s basic structure prior to painting.
AI software is capable of capturing defects of the vehicle in the BIW assembly line. The process scans images in real-time as the vehicle body passes through a vehicle body scanning tunnel that is equipped with Proprietary Lighting Design and multiple-camera setup.

APPLICATION  EXAMPLE  #4

AUTOMATED NON-DESTRUCTIVE TEST (NDT) CELL
FOR POWDER METALLURGY INDUSTRY

Powder metallurgy comes with a lot of advantages in terms of producing customized automotive parts in high volume to save cost. But the disadvantage is that powder metallurgy parts are prone to be affected by defects.
AI software can be easily embedded in all NDT cells to sort defect types and reject defects from the manufacturing line with fast cycle time; and run on a recipe driven based platform to inspect parts with similar size and shape without changover toolings.