MENU

Continental shifts focus for truck electronics to cloud, AI

Continental shifts focus for truck electronics to cloud, AI

Technology News |
By Christoph Hammerschmidt



The “electronic horizon”, a driver assistance system that adapts the driving strategy to the topology and thus reduces the fuel consumption of trucks, has been around for some time. Now Continental is connecting the eHorizon with the cloud to tap even more savings potential. The Dynamic eHorizon will enable commercial vehicle drivers to optimize their driving strategy by providing real-time traffic data from the cloud. This enables truck drivers to save fuel even in slow-flowing traffic – up to two percent more compared to their predecessor, the static eHorizon. Test operation of the sensor system has already started.

Since 2012, Continental’s static eHorizon has saved an estimated 830 million liters of diesel, or more than a billion euros, Continental claims In the dynamic version, as extensive simulations show, up to two percent more fuel savings will be possible in the future compared to the established technology. “The dynamic eHorizon takes into account not only the topology, but also events that are just taking place where the vehicle is moving on its route,” explains Dr. Michael Ruf, Head of Continental’s Commercial Vehicles & Aftermarket Business Unit.

The real-time data on the traffic flow are determined to provide an accuracy of 250 meters in longitudinal direction and being updated every 60 seconds. In the near future, the accuracy of the system will be further increased. Continental merges the data on its own backend with the route data and sends it via the mobile network to the control unit of the dynamic eHorizon.


The movement data of the vehicles ahead are decisive for recognition: The system compares the speed driven with the speeds of the vehicles in front of it. The dynamic eHorizon then distributes the traffic data for the next few kilometers to the powertrain control units. If the system detects slow-moving traffic or congestion, it initiates rolling out, braking or downshifting based on predictive algorithms.

Another technology project at Continental is the right-turn assistant for trucks. The system is designed to prevent accidents involving pedestrians and cyclists when turning right – a type of accident that particularly often causes serious injuries or death to weaker road users. Although radar-based right-turn assistants are already on the market, Continental’s technicians want to achieve a higher degree of significance and thus more reliable operation. Their recipe: Use of object recognition by camera, including gesture recognition.

The next generation will initially supplement the radar image with camera-based object recognition. The image processing algorithms recognize, for example, special features of the traffic infrastructure such as kerbs. This enables the system to classify situations even better and provide more targeted warnings.


The next but one generation of these assistance systems is already in work in Continental’s Computer Vision Lab. AI methods are then also used there to further improve accuracy. The system uses machine learning, neural networks and imaging techniques to detect approaching cyclists or pedestrians much earlier or to anticipate their intentions. Artificial Intelligence also tells the right-turn assistant in which direction a pedestrian or cyclist is likely to move. This should prevent dangerous situations from arising in the first place. The key point here is to make the technology so reliable that it can detect dangerous situations quickly and reliably. At the same time, it must not be set too “sharply”, because false positives would endanger the acceptance of this technology. Continental’s development strategy therefore includes the use of gesture recognition to predict the intentions of a pedestrian or cyclist as reliably as possible. This already works for prototypes, but a lot of development work is still needed: According to Continental’s planning, it will be another five to seven years before the system is ready for the market.

Related articles:

Pedestrian Protection Assistance Systems – Meeting a testing challenge

Continental integrates local weather data into eHorizon

Bosch and Continental buy into virtual map company HERE

 

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News

Share:

Linked Articles
10s