Automated driving in cities

Bosch and Daimler select Nvidia AI platform

  • Bosch and Daimler to source Drive Pegasus platform AI processors and software supplied by the U.S. AI computing company Nvidia
  • Systems architectures for fully automated and driverless urban driving must be versatile, redundant, and fail-operational.
  • Bosch and Daimler machine-learning methods will generate vehicle-driving algorithms.
  • ECU network for automated urban driving handles hundreds of trillion operations per second.

Stuttgart – Automated vehicles are complex computers on wheels. And they need even more computing power if they are to negotiate city traffic automatically, with input sourced from an array of disparate surround sensors. In their alliance to put highly automated and driverless vehicles on urban streets, Bosch and Daimler have specified the computing power needed for their prospective system. The two companies have selected and signed an agreement with the U.S. AI computing company Nvidia as a supplier for the artificial intelligence (AI) platform they need. AI is an important building block in fully automated and driverless vehicles’ network of several individual ECUs (Electronic control units). Under this contract, Nvidia will provide its Drive Pegasus platform powered by high performance AI automotive processors along with system software that will process the vehicle-driving algorithms generated by Bosch and Daimler using machine-learning methods. As a result, the ECU network will reach a computing capacity of hundreds of trillion operations per second. This is akin to the performance delivered by at least six synchronized, highly advanced deskside computer workstations. Bosch and Daimler will also be able to tap Nvidia’s expertise to help develop the platform.

Sensor data processed within fractions of a second
A versatile, redundant, and fail-operational systems architecture is needed to make automated driving in cities a reality. The performance bar for the networked ECUs is no lower, as navigating city traffic is a tremendous amount of work. This network handles all the information gathered and transmitted by disparate radar, video, lidar, and ultrasonic sensors. Just one video sensor, such as Bosch’s stereo video camera, generates 100 gigabytes of data in just one kilometer. The ECU network combines data sourced from all the surround sensors in a process called sensor fusion. Within fractions of a second, it assesses this information and plans the trajectory of the vehicle. This is as fast as the sensation of touch that needs between 20 and 500 milliseconds to reach the human brain. Bosch and Daimler bring many years of experience to the development of functional safety systems. To achieve maximum safety and reliability, the necessary computing operations are done by a number of circuits in parallel. In the unlikely event of a malfunction, the results of these parallel calculations can be accessed in a flash.


ECU network to be integrated into battery cells’ cooling circuit
The high computing capacity and the huge number of operations to be performed mean that the ECU network needs to be cooled. Bosch and Daimler developed an efficient concept based on liquid cooling. In this jointly developed system for highly automated and driverless driving in cities, Mercedes-Benz intends to deploy battery-powered vehicles. These cars have a cooling system on board, so engineers can make the most of this legacy technology by integrating the ECU network into the battery cells’ advanced cooling circuit.

(Source: Bosch Media)

Key elements for autonomous driving functions

Information from ZF:

  • Developing new and extremely high-performance assistance solutions
  • New mobility concepts
  • Vision Zero Accidents

Instead of driving the vehicle, you can opt to do something else: That is the vision of autonomous driving. ZF systems are already coming close to making this vision a reality.

https://www.zf.com/corporate/en_de/products/technologietrends/autonomous_driving/autonomous_driving.html?pk_campaign=20180426-ZF_NorthAmerica_ZF&pk_source=SAE%20Autonomous%20Vehicle%20eNewsletter&pk_medium=E-Mail&pk_content=49-728×90-Sudoku_EN

Car data and who has access?

There is much discussion around at the moment about access to car data and talk of the DLC being removed.

There are genuine arguments on all sides of course but the danger of unauthorised access is very real – particularly as we automate the driving further.

Here is some useful information from http://cardatafacts.eu/vehicle-data-available-service-providers/ which is part of www.acea.be

Interested service providers will be able to access the vehicle data they need through a secure remote server, on the basis of a contract with the vehicle manufacturer.

In addition, independently-managed neutral servers can be set up to make vehicle data readily available to interested third parties without the need to sign a contract with the manufacturer of a car, van, truck or bus. These servers are totally ‘neutral’, meaning that they are neither operated nor financed by the manufacturers but by an independent party. Of course, these neutral server operators are required to implement state of the art security and data protection measures.

Various companies have already shown an interest in setting up such independently-managed servers. IBM, for example, recently launched a service to make vehicle data accessible through their cloud platform to parties that want to develop new and innovative services.

The neutral server will also facilitate data access, in particular for small and medium-sized companies, by offering multi-brand data access on one server, rather than obliging them to use multiple servers of individual manufacturers.

Moreover, the neutral server ensures customer choice. With a neutral server, vehicle users are free to obtain services from the vehicle manufacturer, his network of authorised repairers or any other service provider of their choice.

Service providers can have fair and reasonable access to the data they need to offer their services to vehicle users. That includes independent repair shops, fleet operators, insurance companies, etc. Any information that is available to the vehicle manufacturer’s network of authorised repairers will be made available on the same conditions to independent third parties that offer competing services: the same type, amount and quality of data, at the same time, at the same price.

This concept for the transfer of vehicle-generated data ensures access in a fully transparent and anonymised manner. That is, the neutral server enables service providers (as well as the exact services they offer) to remain unknown to the vehicle manufacturer. Thus, it contributes to innovation and allows fair and open competition.

Cycle recognition and emergency braking

Introduction

 

Another narrow escape: a cyclist appears as if out of nowhere and suddenly crosses the road. Distracted by the search for somewhere to park, the driver is powerless to avert what appears to be an inevitable disaster. Yet Bosch’s new emergency braking system with cyclist detection prevents any serious consequences, automatically bringing the car to a full stop from 40 kph. Everyone makes it through the incident, shaken but unharmed. As soon as the emergency braking system’s radar or video sensor detects an imminent collision, the Bosch iBooster initiates full braking in just 190 milliseconds – less time than it takes to blink twice. “Driver assistance systems are the next step along the path toward accident-free driving,” says Bosch board of management member Dr. Dirk Hoheisel. “These electronic assistants are always vigilant and, in emergencies, they respond more quickly than people can. They provide support just where drivers need it – in busy city traffic.” Emergency braking systems are one of the most useful assistance systems, particularly when it comes to responding to cyclists and pedestrians, the most vulnerable of road users.

 

More protection where most needed

In Germany, bicycles are involved in one-fourth of all accidents resulting in personal injury. According to the German Federal Statistics Office, 393 people were killed in such accidents in 2016 alone – roughly 12 percent of the country’s total road fatalities. Some two-thirds of these accidents involve a car. Equipping every car in Germany with an emergency braking system that can detect cyclists would prevent almost half (43 percent) the bicycle/motor vehicle accidents that result in personal injury, or at least mitigate their severity. “An emergency braking assistant may reduce braking distance by the few crucial centimeters that can mean the difference between life and death,” says Gerhard Steiger, president of Bosch’s Chassis Systems Control division. The European New Car Assessment Program, or Euro NCAP, has also recognized the importance of emergency braking systems for road safety. Starting in 2018, the consumer protection association’s star rating system will include emergency braking with cyclist detection. Emergency braking systems with pedestrian detection have been part of the rating system since 2016.

Electronic assistants growing in popularity

In light of rising volumes of road traffic, driver assistance systems offer the full package – and hold the key to increased road safety. They keep cars in their lanes, warn of obstacles in the blind spot when changing lanes, provide support for pulling into and out of parking spots, and help maintain following distance, to name just a few examples. Bosch is constantly honing the technology behind these driver assistance systems: sensors supply increasingly precise images of the car’s surroundings, and their interaction with actuators, such as braking and steering, is steadily becoming faster and more efficient. In this way, driver assistance systems are not only preparing the path toward automated driving, but are already delivering stress-free and relaxed driving. No wonder, then, that the spread of electronic assistants is picking up. A Bosch survey found that half of all new cars (52 percent) in Germany have at least one driver assistance system on board. The trend is toward consolidating multiple assistance functions on one sensor, as demonstrated by car exit warning, a new function developed by Bosch.

Radar offers a constant over-the-shoulder view

Bosch’s rear mid-range radar sensors, which monitor lane changes on the freeway, can also keep city drivers from making a dangerous mistake: after parallel parking at the curb, drivers often get out of their cars right away – without looking over their shoulder. This has led to countless cyclists getting painfully up close and personal with car doors as they are knocked unceremoniously to the pavement. But Bosch’s car exit warning can help. It is active for all car doors and warns the occupants – even several minutes after the ignition has been turned off – before they carelessly get out of the vehicle. Mounted to the left and right of the rear of the car, the Bosch sensors monitor traffic. Within a 20-meter radius, the sensors can detect other road users who are approaching from the rear, or who are already to the side or rear of the car, and promptly warn the driver before they open their door.

 

(Source: Bosch Media)

Tesla new autopilot features

Every Tesla vehicle comes standard with full self-driving hardware – enabling the driving experience to be substantially safer than that of a human driver. With their most recent software update, you can now experience our most advanced safety features, including Traffic Aware Cruise Control, Autosteer up to 90mph, Automatic Emergency Braking, and Side-Collision Avoidance.

 https://www.tesla.com/en_GB/videos/autopilot-self-driving-hardware-neighborhood-short

(Source: Tesla)

Automated mobility – Bosch

Automated driving impacts the entire car: its powertrain, brakes, steering, display instruments, navigation, and sensors, as well as connectivity inside and outside the vehicle. The key to success is an in-depth understanding of all vehicle systems. Few automotive suppliers worldwide have as much knowledge in this area as Bosch, in part because the supplier of technology and services manufactures most of the components needed for automated driving listed below:

Connected Horizon: Automated vehicles rely on environmental information – information that goes beyond what sensors can gather. For instance, they need real-time traffic data on congestion and accidents. This can be achieved only by connecting the vehicle to a server, for which Bosch developed its Connected Horizon solution. This system enables a dynamic preview of the upcoming route and corresponding adjustments to driving strategy. Connected Horizon is what allows automated vehicles to think ahead. This is beneficial for the comfort and safety of the driving experience. For instance, connected vehicles are warned in advance of danger spots before a blind bend or hilltop and can ease off the accelerator in preparation.

Electric steering: Fail-safe, electric power steering is a key technology for automated driving. Even in fall-back mode, fail-operational capability allows drivers and automated cars to continue using essential steering functions while maintaining about 50 percent electric steering support in the rare case of a malfunction. This technology will enable automakers to comply with the safety requirements as proposed in the Federal Automated Vehicles Policy documents from the U.S. Department of Transportation and National Traffic Highway Safety Association, for example.

ESP: The electronic stability program also plays a key role when it comes to automated driving. Delegating responsibility for driving to the vehicle places particular demands on safety-critical systems such as the brakes. To retain maximum control over these systems in the event of the failure, redundancy must be built into the system as a safeguard. In this instance, the ESP brake control system and iBooster electromechanical brake booster (see below) can independently brake the vehicle without the driver having to intervene. Bosch offers ESP as a modular concept that offers the right system for all circumstances and requirements.

HMI: Automated driving will change the human-machine interface, and calls for modern concepts for communication between car and driver. The driver must be able to intuitively understand and use the system. With its innovative display instruments, Bosch is already offering promising solutions in this area as well: the TFT instrument cluster, for instance, offers maximum flexibility in processing combined with brilliant clarity. By using head-up displays, Bosch puts information such as speed, navigation prompts, and warnings directly in the driver’s field of view. This information is superimposed on the vehicle’s surroundings in such a way that the two seem to blend seamlessly at a distance of around two meters ahead of the vehicle.

iBooster: With the iBooster, Bosch has developed a vacuum-independent, electromechanical brake booster that meets the requirements for modern braking systems. It can be used in all powertrain concepts and is especially well suited for hybrid and electric vehicles. In the iBooster, the actuation of the brake pedal is recorded by the built-in pedal-travel sensor and transmitted to the control unit. The control unit calculates the triggering signal for the electric motor, which uses a two-stage transmission to convert its torque into the required power assistance. In a standard master cylinder, the power provided by the booster is transformed into hydraulic pressure.

Maps: Without high-resolution, up-to-date maps, there can be no automated driving. The maps provide vehicles with information about changing traffic situations, such as traffic jams or construction, that fall outside the area on-board sensors can monitor. Bosch’s radar and video sensors capture and transmit important real-time traffic data for the creation of high-resolution maps for automated driving.

Lidar sensor: In addition to radar, video, and ultrasonic sensors, Bosch also uses lidar sensors in its automated test vehicles. The various sensor principles complement each other very well and combine data to ensure reliable environment recognition. Automated vehicles use this data to derive their driving strategy. Bosch views lidar sensors as an important addition to its portfolio.

 

Radar sensor: As one of several sensor principles, radar sensors provide important 360-degree information about their surroundings within a distance of up to 250 meters for automated vehicles. A radar sensor’s main task is to detect objects and to measure their speed and position relative to the movement of the vehicle. Furthermore, Bosch radar sensors send frequency-modulated radar waves measuring between 76 and 77 GHz via a transmitting antenna. These waves are reflected by objects in front of the vehicle. The relative speed and distance of objects are measured using the Doppler effect and the delay generated by the frequency shifts between the emitted and received signal. Comparing the amplitude and phase of the measured radar signals makes it possible to draw a conclusion about the position of the object.

Ultrasonic sensor: Ultrasonic sensors are needed in automated driving, primarily for close-range environment recognition of up to 6 meters and at low speeds, such as during parking. The sensors employ the sonar technique, which bats, for example, also use in navigation. They emit short ultrasound signals that are reflected by obstacles. The echoes are registered by the sensors and analysed by a central control unit.

Video sensor: With a 3D measurement range of over 50 meters, the Bosch stereo video camera provides important optical information about the vehicle’s surroundings. Each of the two highly sensitive image sensors, equipped with colour recognition and complementary metal oxide semiconductor (CMOS) technology, has a resolution of 1280 by 960 megapixels and is capable of processing extreme contrasts. The distance between the optical axes of the two lenses is just 12 centimetres. The stereo video camera captures objects spatially and calculates their distance, plus it identifies clear spaces. The information from the sensor is combined with data from other sensor principles to generate a model of the surroundings for automated vehicles.

Source: Bosch Media, Bosch Pictures

“Just driving” was yesterday – the personal assistant is tomorrow

Bosch’s new show car shows how quickly the future of driving is becoming a reality

  • Connected, automated, and personalized: Bosch has a new take on mobility and is turning the car into people’s third living space
  • New user interfaces ensure more security, more comfort, and fewer distractions when driving
  • Cars are becoming personal assistants on four wheels

Stuttgart – My home, my workplace, my car: connectivity is turning cars into a third living space alongside people’s own home and their office. Bosch is showing what that actually means, and what it will be like to drive a car in the future, with its new show car. It offers intuitive operation and is always online, connected with its surroundings, and driving itself. “The connectivity of cars with their surroundings and with the internet is a key challenge for future mobility,” says Dr. Dirk Hoheisel, member of the board of management of Robert Bosch GmbH. Automated and connected functions in cars not only make each journey safer and more comfortable, they also turn the car into a truly personal assistant. “In this way, we are making connectivity a personal experience and giving people more time for actual living, even while driving their car,” Hoheisel says.

Intelligent display and user interfaces

More individuality and easier operation become apparent as soon as you get into the show car. The driver monitor camera recognizes the driver and adjusts the steering wheel, mirror, and temperature accordingly. In fact, as if by magic, the car also sets the colour scheme of the display and automatically loads appointments, favourite music, the latest podcasts, and the navigation destination that the driver programmed while still at the kitchen table. The camera is always alert during driving, too, especially when the driver’s eyes get a little heavy. It detects fatigue and microsleep at the wheel, both of which are often the cause of serious accidents. It is usually possible to spot the onset of these early on from movements of the eyelids. The system determines the driver’s ability to concentrate, or degree of tiredness, and issues a warning if necessary. This makes driving even safer. What is more, the driver tiredness detection system constantly monitors the driver’s steering behaviour so it can intervene directly in the event of abrupt movements.

The human machine interface (HMI) turns cars into personal assistants on four wheels. This interface between people and vehicles provides drivers with important information when it is needed and is an attentive alert companion in every situation. In the future, thanks to more personalized communication, automated and connected functions will offer intuitive, comfortable, and safe operation, and drivers will be able to set them to meet their personal requirements – whether in a traffic jam, in urban traffic, or on a family outing. To this end, the show car presents gesture control with haptic feedback. It uses ultrasonic sensors that produce a noticeable resistance whenever the driver performs a gesture in precisely the area that the camera records. This makes gesture control even easier to use and less distracting for drivers, since they can change the information on the display, accept phone calls, or call up a new playlist without touching it. An innovative touch display in the show car also makes it safer and more convenient to use fingertip control. The display provides a haptic response by vibrating each time the driver’s fingertips touch it. This means drivers can sense different structures that feel like real buttons on what is in fact a flat surface. That way, they can easily find the desired function on the display, for instance to adjust the volume of the music, without looking away from the road.

Mobility with smart connectivity: Cars are turning into people’s third living space

The show car also demonstrates how cars are turning into people’s third living space thanks to automation and connectivity. According to Bosch’s “Connected car effect 2025” study, automated driving could enable people who drive a lot to make better use of some 100 hours of their time each year. Once the car detects that automated driving is possible and the driver agrees to hand over control, the car takes over – safely and smoothly. Since the show car is an active part of the internet of things, drivers can carry their digital lives over into their car; perhaps sending e-mails to the office colleagues or video chatting with friends. All this is possible in the time automated driving saves. Flexible display concepts really come into their own here. Drivers can simply gesture to seamlessly switch like magic between various displays of e-mails, chats, videos, and automated and connected functions.

Connected with the smart home, the repair shop, and the whole world

What about planning your evening meal when on the road? Connectivity can help here, too – this time with the smart home. Mykie, the Bosch kitchen assistant, can suggest recipes online in the car. A glance from the car into the connected refrigerator will show whether the necessary ingredients are ready at home. Connectivity between cars and smart homes comes into play even before the journey starts: as soon as drivers enter the car, a display shows them the status of their own home. Has a window still been left open? Is the door locked? It takes just a gesture or a fingertip on the display to automatically lock the doors and monitor the status at home. Moreover, the connected car is also linked to the repair shop. It notifies drivers when an inspection is due, it schedules an appointment at the repair shop upon request, and it can ensure the necessary spare parts are in stock when it gets there. This level of comfort extends to parking: in Bosch’s community-based parking service, cars use the sensors in parking assistants to report available curbside spaces. This information is sent via the cloud to a digital parking map and provided to other vehicles.

Source: Bosch Media

Datacentres: In the driving seat of the connected car revolution

Here is an interesting article by

(16 December 2016, 11:11 a.m.) on the IOT site:

http://www.iottechnews.com/news/2016/dec/16/datacentres-driving-seat-connected-car-revolution/

When I started driving, cars were generating very little data. They got you from A to B without the addition of gadgets or gizmos. Connected cars as we know them today were certainly not a thing.

Today many vehicles are computers in their own right, connected to the Internet and data is flooding in. In fact, it’s estimated that a single connected car uploads 25GB of data to the cloud per hour.

(c)iStock/aleksle

With a quarter of a billion smart vehicles set to be on the road by 2020, that’s over 6 billion GBs every 60 minutes.

Such vast amounts of data are only going to continue growing in the years to come, putting the automotive industry in a leading position within the Internet of Things (IoT).

But at the same time a growing number of challenges and pressures are becoming apparent – namely the need to process, analyse and store all this new information.

As a result, datacentres are fast becoming the solution to the automotive sector’s rapid data growth, but how exactly are these data halls driving the connected car revolution forward?

From connected cars to autonomous autos

For the past few years, connected cars have been the hype of the sector.

By ‘connected’, we mean vehicles that have access to the internet in some form; cars that are often spotted with sensors that enable machine to machine (M2M) and machine to human (M2H) communication. As already noted, this level of connectivity generates substantial data sets.

The industry is continuing to innovate rapidly, and before connected cars even become commonplace, conversations are shifting to autonomous (or self-driving) vehicles – the futuristic Hollywood vision realised.

Here we’re talking about vehicles that operate without a human driver. While this could well give rise to many transportation efficiencies (reduced driving costs, improved convenience etc.) it will also undoubtedly bring about a more drastic automotive data revolution.

If one connected car today generates 25 GB of data an hour, one autonomous car in the future is likely to generate ten times that information.

On top of all the data a connected car generates, self-driving vehicles will have to be truly intelligent – learning how to their ‘drivers’ like to drive, sensing the physical environment around them, broadcasting location data and interacting with other vehicles and objects to traverse the roads safely.

By producing data on data in this way, autonomous cars will require even quicker analysis and bring entirely new elements of machine learning to the mix.

Which means beyond M2M/M2H communication we must also consider vehicle to vehicle (V2V), vehicle to everything (V2X), vehicle to infrastructure (V2I) vehicle to person (V2P) and vice versa (P2V).

Driving datacentre demand

The resulting complexity and scale of automotive data sets means more and more automotive giants are recognising the need for complex computing to drive their businesses (and vehicles) forward.

HPC – and the datacentre industry as a whole – sits in the driving seat of the intelligent automotive revolution

In turn, this has resulted in an exponential growth in the number of customers from the automotive industry turning to external data centre providers to meet their Big Data and High Performance Computing (HPC) demands.

The need for scalable, secure HPC datacentre solutions is therefore being felt keenly. For many auto-companies, these kind of data hubs are not necessarily those on their doorstep, and IT decision makers are looking to colocation datacentre providers to support their HPC operations, by supplementing compute capacity and improving operational costs.

In order to support the rapid innovation the automotive industry is showing at present, such campuses must present an ‘HPC-ready’ solution – offering the expertise to support the management of information loads as quickly, efficiently and successfully as the automotive experts that have been handling complex vehicle data for decades.

Innovating in Iceland

More often than not, these are remote facilities with the power infrastructure, resiliency levels and computing resources needed to process HPC loads cost-effectively. Moving automotive HPC workloads to campuses with inherent HPC-ready capability gives automotive manufacturers the medium and high power computing density required at significantly lower energy costs.

Ultimately that enables the ability to gain more insight from more data, and moves us closer to the benefits of autonomous driving.

A number of automotive leaders have recognised these benefits, and are already reaping the rewards. One such manufacturer is Volkswagen, which recently announced the migration of one megawatt of compute-intensive data applications to Verne Global’s Icelandic campus in order to support on-going vehicle and automotive tech developments.

Likewise, BMW is a well-established forward-thinker in this area, having run portions of its HPC operations – those responsible for the iconic i-series (i3/i8) vehicles, and for conducting simulations and computer-aided design (CAD) – from the same campus since 2012.

These automotive leaders consider Iceland an optimal location for their HPC clusters – not only for the energy and cost efficiencies it delivers, but the opportunity it allows them to shift their focus from time-intensive management of the technical compute requirements of their day-to-day work to what’s really important: continued automotive innovation.

Even so, wherever automotive data is stored, analysed and understood one thing is for sure: HPC – and the datacentre industry as a whole – sits in the driving seat of the intelligent automotive revolution.

It will advance our understanding of auto-tech, smarten our driving behaviours and ultimately carve a path to the coveted driverless and connected car technologies that will radically change the way we travel into the future.