ADAS: A Bridge to Self-Driving Cars

Bridge to Self-Driving Cars

The Problem with Our Driving is… We Are Human

The invention of the automobile has impacted us like nothing else. It allows us great independence in the way we live our lives, making it easy for people to go wherever they want wherever they want. However, that independence comes with a price. Since the automobile is still operated by a human, we have to deal with the consequences of human mistakes, in the form of accidents on a daily basis. Nearly 1.3 million people die in road crashes each year, on average 3,287 deaths a day. An additional 20-50 million are injured or disabled. Everyone has heard the future is self-driving cars, but what can we do in the present to reduce these numbers? The answer comes in the form of Advanced Driver Assistance Systems (ADAS) and VIA Mobile360.

ADAS History

ADAS in some forms is not a new idea. Ever since the Model T, manufacturers have had the goal of safety and driver ease in mind. From the Curb Feelers of the 1950’s to the Anti-Lock Brakes of the 1980’s, there have been small incremental increases in ADAS technologies. Even these small changes have allowed a marked decrease in accidents and injuries, but advances in technology such as VIA Mobile360 ADAS will finally make drivers as accident-free as humans can get.

The Sensor Revolution

Computer Vision is the practice of giving machines knowledge of their physical surrounding world through sensors. These sensors can be anything from the simple Infrared sensor, which is very inexpensive but is only good at judging distances, to LIDAR systems, which are extremely accurate but also very expensive. Each of these sensors has its place in a lot of applications, but the overall best sensor for ADAS is the camera. Cameras offer the best combination of low cost and high quality. Combined with the help of Artificial Intelligence, they can alert the driver of possible danger.

The A.I. Revolution

In the past, Computer Vision was a very fragile and complicated task requiring a specific tailored algorithm to analyze pixels. These algorithms were not flexible and had to be used in a specific case and were very susceptible to rotation and lighting. For instance, if you wanted to recognize a car, you would have to write an algorithm for different car models, in multiple lighting conditions, from almost every angle to be accurate. This has all changed with Artificial Intelligence and Machine Learning. Now, there are no specialty algorithms needed to identify a car. In fact, the computer teaches itself what a car looks like by learning from many images of cars. This allows for great flexibility and accuracy.

The Hardware Revolution

The GPU or Graphics Processing Unit is a specialized processing unit originally intended for acceleration of graphics in video games. The GPU is far superior at certain mathematical operations than the standard CPU. Matrix math is the power behind graphics processing and, luckily, it is also used in Deep Learning. Recent developments in the speed of hardware GPUs have allowed Computer Vision to take advantage of Deep Learning Networks (a subset of A.I.) that help to mitigate the issues experienced with standard Computer Vision algorithms like the scenario mentioned above. For the VIA Mobile360 product line, we are adopting the state-of-the-art Qualcomm® Snapdragon™ 820E quad-core processor with an Embedded Adreno™ 530 GPU that allows for excellent power and efficiency.

The ADAS Revolution

With its leading-edge graphics, AI, and Computer Vision capabilities, VIA Mobile360 ADAS makes driving smarter and safer by supporting features like:

mobile360 ADAS Sample Kit

Lane Departure Warning

Have you ever been distracted while driving and suddenly drift into the other lane without noticing or failed to signal? VIA Mobile 360 ADAS will alert you when accidently go out of your lane.

Forward Collision Warning

When driving sometimes we forget about what the safe driving distance is to the car in front of us. Now this is calculated many times per second and we can be warned if we are too close.

Blind Spot Detection

Using the two automotive-grade FOV 50 cameras mounted on the side of the vehicle, VIA Mobile 360 ADAS has a constant view of your blind spots and will notify you when a possible unseen vehicle is in the way.

Pedestrian Detection

Being human, we sometimes get “Tunnel Vision” while driving and only look for other cars. Luckily machines don’t have this problem. Using Deep Learning, VIA Mobile360 ADAS can alert you of pedestrians when driving.

Vehicle Detection

With a total of four automotive-grade FOV 50 cameras, VIA Mobile360 ADAS can detect vehicles and judge their distance everywhere around the car.

Classification for special vehicles, motorcycles, bicycles

VIA Mobile360 ADAS is trained to recognize special vehicles such as fire trucks, police cars, motorcycles, and bicycles – keeping you safe and possibly avoiding costly traffic violations.

Speed Limit Detection

The front-facing camera constantly scans to detect speed limit signs and can alert you when you are speeding.

Rear-end Collision Avoidance

The rear facing camera recognizes vehicles behind you and alerts you of them following too close in order to prevent you from being rear-ended.

The system also features VIA Mobile360 E-Track, a cloud portal that enables fleet owners to collect and organize vehicle and driver data for real-time vehicle tracking, event and data recording, and asset management. Customizable event triggers keep managers informed of critical situations so that they can respond to immediately, while a user-friendly graphical interface makes analysis of trip reports a breeze.

VIA Mobile360 ADAS is an ideal solution for a wide variety of applications, including

  • Armored Cars & Secure Freight Shipping
  • Buses and Coaches
  • Refuse Collection
  • Freight Transportation & Delivery
  • Construction, Industrial, & Mining
  • Agriculture & Forestry

In the future, it is a certainty that all forms of transportation will be completely autonomous. But, this doesn’t mean you have to continue driving in the 20th century while you wait for that to happen. Why wouldn’t you protect your company’s assets and employee’s safety today by stepping toward the future with VIA Mobile360 ADAS and bridge the gap to self-driving cars?  For more information about how VIA Mobile360 ADAS can help your business, please contact us.


About the author:

Jason Lee Gillikin has 6 years experience in the IoT field helping to develop better and smarter products.  He holds degrees in Physics and Chemistry with a Masters Degree in Computer Science and Information Systems from the University of North Carolina.   Currently he is the Business Development Manager at VIA Technologies helping bring these products to the market. You can contact Jason on LinkedIn or by email.

VIA Technologies, Inc.