How BMW is using Artificial Intelligence
--
A German car manufacturer well known for the best in class designing, manufacturing and performance, BMW is a well known name in the car manufacturing business across the globe. From BMW 2 Series Gran Coupe to BMW M8 Coupe, it has a wide variety of cars for every budget. It designs and builds some of the most advanced, high-tech automobiles in the market.
In this age of Artificial Intelligence, BMW has already blended AI with their work culture and manufacturing to bring out the best automobiles of tomorrow. There are several sectors where BMW is using cutting edge AI technology, some of which are mentioned below.
Reducing Monotonous Human Tasks in Manufacturing and Production
BMW’s AI Intelligence model relieves workers from monotonous tasks such as checking whether the warning light has been correctly positioned on the car’s boot etc. by performing the same using a camera and a self learning software that relays the camera’s live information and compares it to hundreds of stored information in the blink of an eye.
AI enabled Quality Control system
Despite BMW’s state-of-the-art filtration technology, dust particles in paint are still a possibility which could significantly impact the painted surface. This could compromise the paint surface finesse as well as the quality of the product for which BMW is known for. To tackle this, BMW uses the state of the art Artificial Intelligence system to increase the paint shop quality. This is achieved by an automatic surface inspection in the paint shop which consists of gathering data to be used as part of a high-level data base for dust particle analysis. The AI system then evaluates live data from dust particle sensors in paint booths and dryers with the help of this database.
BMW and NVIDIA partnership
On May 14th 2020 at NVIDIA’s GTC Conference, NVIDIA announced that BMW has chosen to partner with NVIDIA Isaac robot platform to improve its automotive factories .NVIDIA assisted in manufacturing 5 AI-enabled robots designed to improve workflow; Smart Transport Robot, SplitBot, PickBot, PlaceBot and SortBot. The main task of these robots is to sort, move and organise boxes and pallets. The robots are trained using real and simulated data using NVIDIA’s GPU. The data is then used on NVIDIA DGX system.
Using Big Data and Artificial Intelligence to create self driving cars
BMW believes that the cars of tomorrow will drive themselves, instead of relying on humans being at the wheel. The company has said it plans for its cars to deliver full “level 5” autonomy, which is the highest level determined by the US department of Transport by 2021.
They also planned to gather location based data from cars to train the self driving cars of the future. This includes machine data such as braking force, wiper and headlight use, video data from on-board cameras, and GPS data. These data can then be fed to the computer vision provided by the Intel’s Mobileye Technology for training . These cars won’t drive themselves or function without a human driver, but, as autonomy level 3 and 4 vehicles, they can perform many driving maneuvers and procedures unaided.