in , ,

Here’s how Continental is trying to make roads safer from autonomous cars

Here’s how Continental is trying to make roads safer from autonomous cars
Alexander Klotz, Head of Technical Center India, Continental Automotive India

Accident detection is one of the most crucial components of automotive safety. Based on the combination of historical and current data, Continental has developed a mechanism which can detect if a crash is about to happen and then take the necessary steps to avoid it.

“We are using AI for developing test cases and scenarios. Developing a safe and reliable vehicle needs millions of test runs. For instance, we cannot crash millions of cars for crash detection before reaching a level of confidence. AI/ ML helps in analyzing historical data and produces simulations and different test cases. Apart from that, AI systems are used in other functions like object detection; they create simulations and automatically compare outcomes to produce reliable systems,” Alexander Klotz, Head of Technical Center India, Continental Automotive India, told ETCIO.

To successfully shape this transformation and advance its technological leadership, Continental has invested in setting up its supercomputer for Artificial Intelligence (AI), powered by NVIDIA InfiniBand-connected DGX systems. It has been operating from a datacenter in Frankfurt am Main, Germany, since the beginning of 2020 and is offering computing power and storage to developers in locations worldwide. AI enhances advanced driver assistance systems, makes mobility smarter and safer, and accelerates autonomous driving development.

“AI enhances ADAS (Advanced Driver Assistance Systems) and makes mobility smarter and safer. Enhanced ADAS systems are bringing autonomous driving within reach of mid-term business plans. The supercomputer is currently used to train neural networks for 70% and virtual testing for 30%,” Klotz added.

The data used for training neural networks comes mainly from the Continental test vehicle fleet. Currently, they drive around 15,000 test kilometers each day, collecting around 100 terabytes of data – equivalent to 50,000 hours of movies.

Already, the recorded data can be used to train new systems by being replayed and thus simulating physical test drives. With the supercomputer, data is generated synthetically, a highly computing power consuming use case that allows their systems to learn from travelling virtually through a simulated environment.

Automated and assisted driving continues to be one of Continental’s growth areas, and the German Automotive company is continuously innovating in the domain. Some of the ADAS technologies with high potential include hood visibility, emergency braking system and more.

Transparent Hood

The “transparent hood” function makes the area under the hood visible and allows the driver to see terrain and obstacles that would otherwise not be visible.

“Transparent Hood is based on our Surround View system, consisting of four satellite cameras and an electronic control unit. An intelligent image processing algorithm reconstructs the image below the vehicle and inserts this image exactly into the surround-view displayed to the driver. The technology fuses ultrasonic distance information with Surround-view cameras to provide a 3D perspective, 360° around the vehicle. It supports views both near the vehicle, as well as in the medium distance,” Klotz highlighted.

Other than this car safety feature which allows the driver to gain visibility through the vehicle’s hood, the automotive manufacturer has also invested in new age technology to enhance the right turn assist system.

The Right-Turn Assist system for passenger cars makes turning-off significantly safer. It provides welcome support for car drivers, especially in the complex situations that sometimes occur in urban traffic. The new generation of radar sensors can detect a cyclist approaching from behind the vehicle on the right-hand side.

“Right-Turn Assist intervenes if the driver wants to turn right when the cyclist is just about to pass along the vehicle’s right-hand side. Drivers are not always able to recognize this risky situation, even when they look over their shoulders. If the radar sensors detect a cyclist in such a case, they transmit a signal to the brakes, and the car stops before it collides with the cyclist. This safety function also protects pedestrians and scooter riders,” Klotz added.

The concept is integral to protecting vulnerable road users (VRUs) such as pedestrians and cyclists. Continental has a model called Intelligent intersection which supports drivers in complex intersection traffic scenarios, like preventing left-turning cars from running head-on into traffic that approaches behind an obstacle. It can also inform drivers are turning right about traffic coming from the left.

Intelligent Intersection

Continental’s Intelligent Intersection is a real-world, end-to-end solution comprising a sensor set, powerful sensor fusion algorithms to generate a comprehensive environmental model. And Dedicated Short-Range Communication (DSRC) to transfer valuable information between the intersection and connected vehicles. The solution makes it possible to warn an oncoming driver about a crossing pedestrian outside the driver’s line of sight. Its data can also control signal changes, streamlining traffic, and reduce emissions and idling time at intersections.

Additionally, Continental’s Intelligent Intersection enables the collection of information and statistics from intersections – often congested, high-incidence zones – which can then be used for traffic flow, safety, and environmental improvements.

“To enhance driver safety while protecting vulnerable road users, we have technologies like Emergency Brake Assist (EBA) and Autonomous Emergency Braking (AEB). EBA compensates for insufficient driver action on brakes. Drivers in emergencies either apply insufficient pressure or release the brake too early. Not only does EBA address this problem, but it can also reduce stopping distances significantly. AEB is a collision avoidance feature that integrates state-of-the-art driver assistance systems with Electronic Stability Control (ESC),” Klotz emphasised.

It is a proactive and predictive driver assistance system that provides emergency braking assistance or brakes autonomously when an accident risk is detected. According to Klotz, this intelligent system aims to avoid collisions with all types of obstacles. Alternatively, if an accident is unavoidable, the system targets the maximum reduction of the impact speed.

Source: cio.economictimes.indiatimes.com

What do you think?

486 points
Upvote Downvote

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Smart cities can speed up autonomous driving

Smart cities can speed up autonomous driving

Driverless cars: what technology do they need apart from being “driverless”?

Driverless cars: what technology do they need apart from being “driverless”?