By Ralph Lauxmann and Andree Hohnn:
Different types of automated driving complement one another to create the seamless mobility of tomorrow
There are four main global drivers that require the innovative engagement of the automotive industry and, with it, the large technology suppliers: The vision of accident-free driving, air quality problems, irritation over time lost in traffic jams, and an increasingly older population. This is where connected automated driving can make an important contribution. But what does this actually mean for a car journey? Which systems and solutions must a supplier have in his portfolio to enable seamless and comfortable individual mobility?
Far too many people are still being killed in road accidents. More than 1.3 million people lose their lives in this way every year. And this is just one of the big challenges facing technology companies for mobility solutions.
Increasing traffic levels also mean that the issue of emissions is becoming more and more important. This applies particularly to areas with dense populations, which will only increase in the future. Some projections predict that in just a few decades possibly half of the world’s population will live in megacities. Mobility must change if we do not want our lives to be completely dependent on face masks and NOX alarms.
These environments are dominated by commuting between home and work and by recreational traffic. These days, however, most drivers spend a lot of their time in congestion. London is already at risk of total gridlock. With a current average speed of 7.8 miles an hour, you can hardly use the term “automotive mobility” in Britain’s capital city.
On the contrary, traffic is practically back to the speed level of the horse and cart as at end of the 19th century. The traffic figures are even worse in megacities in other continents. New solutions are, therefore, required to maintain and increase individual mobility.
Demographic development is another challenge. The proportion of older drivers is increasing. The decision to stay behind the wheel is not always simply a matter of free choice. In rural Japan, for example, people are often forced to continue driving into their old age because they have no other option of getting around. There are already not enoughyounger people living there who can work as taxi drivers.
Conditions in Germany are very different, but after 2030 there will be more people here drawing a pension than in gainful employment. This development will change the demands on mobile technology as well.
Automation – Step 1: Starting From Home
The technology of automated driving (AD) will provide an important contribution to making life easier for the driver, and to creating a more efficient overall traffic flow, with fewer critical situations or accidents. Particular attention must be given to fully automated driving in line with the current SAE/VDA definition Level 4 as well as Level 5, driverless.
Level 4 refers to connected vehicles which are able to assume driving tasks completely in certain driving situations. The driver can then focus on other tasks and use the travel time meaningfully. At Continental’s automated driving project house, the Cruising Chauffeur and Self Driving Car projects utilise a common technology base, as well as numerous other technologies from Continental’s wide range, to achieve the different levels. A hypothetical trip from Hamburg to Munich illustrates which technical conditions have to be satisfied.
Several automation capabilities are required to make this trip safe and comfortable. Let’s assume the driver is starting the journey from home. The first task is to find an arterial road from the city centre – using the driver’s own knowledge or a navigation system – and from there to get onto the correct highway. If the sensor system and vehicle intelligence recognizes the highway situation, the driver is given the option to transfer the driving task to the vehicle (“AD is now possible”). If the driver decides to accept this option, the AD system assumes the role of the “Cruising Chauffeur” for the highway section of the journey (Figure 1).
The technical infrastructure required for this process along the logical chain Sense Plan Act (sensing the environment, action planning for the control units and activating the actuators) may be invisible to the driver, but it is remarkably extensive (Figure 2).
The environment sensor system alone, which is part of the automation process, includes several camera types (mono, stereo, surround view), radar sensors of varying view ranges and range characteristics (long range radar, short range radar) and LiDAR sensors (for example, high-resolution 3D Flash LiDAR). This data is used together with information from vehicle-to-vehicle communication, communication with other motorists, the infrastructure and the cloud, as well as high-resolution map data, to create an extensive model of the driving environment. Based on this so-called environment model, the main participating control units (assisted and automated driving control unit as well as the safety domain control unit) for basic driving and safety functions, such as the minimal risk manoeuvre, can determine and implement the appropriate driving strategy.
The system intervenes in the braking, steering and engine control units to implement the driving strategy during highly automated and autonomous driving. At the same time, all actuators must be designed with built-in redundancy to ensure that the required actuations are still carried out even in the event of a fault. If we use the brake as an example, this means that Continental combines thehighly dynamic MK C1 with the additional module for a redundant brake system, the Highly automated Brake Extension, HBE.
The complexity is even greater at the overall vehicle level: Solutions from all the divisions of Continental work together to provide an integrated, single-source solution. The most important issue for the driver is how to communicate with the Cruising Chauffeur and how the Cruising Chauffeur can respond to the actuation and requests of the person behind the wheel.
Once again, a sensor system is required for this, as it is an important part of the interaction concept between man and machine. The development of the Cruising Chauffeur in accordance with SAE/VDA Level 4 is currently being pursued. Almost all vehicle manufacturers have announced their intention to implement comparable vehicle-based systems.
Technologically, Continental predicts series production of AD Level 4 from 2020. However, this depends on legislation. Development of automated driving at Continental is currently focusing on environmental modeling and algorithm development for the driving strategy.
Automation Step 2: Vehicle Parks Itself
Let us assume that our driver is approaching Greater Munich after a long and relaxed journey – thanks to automated driving on the highway – and will therefore soon reach the end of the Cruising Chauffeur’s control task. The person behind the wheel is now asked in good time to resume driving.
This stage of the journey could lead to a park-and-ride parking lot to ensure maximum comfort and mobility. This is still a sticking point for many drivers today, as the transition from a comfortable passenger car to the next means of transport can feel like a cold shower. There are many reasons for this. You start out by looking for a parking space in a typically large area. Then you might have to walk quite a distance, possibly with luggage and in bad weather, to the next public transport stop. The timings of buses or trains can cause further irritation if they result in long waiting times. Another common problem can be broken ticket machines.
These kinds of experiences cause many drivers to drive directly to their destination, even though many have probably already experienced the stress of driving the last few miles to their destination. This means that even though the car is comfortable for longer journeys, it can feel like quite a liability in city centres, due to a lack of parking space and heavy traffic – welcome to London.
Automation can be a great help in this situation as well, because a vehicle with an automated parking function (Figure 3) fundamentally changes the qualityof this part of the journey: Drivers and passengers entrust the vehicle to a mobility terminal instead of having to look for a free space themselves. At the press of a button, the vehicle autonomously looks for a free space, and then parks and locks itself (Valet Parking). This is an enormous advantage for the operator of the parking lot, because the vehicles can be parked much closer together, making it possible to use the available space much more efficiently.
The vehicle infrastructure for automated parking basically corresponds in type and scope to the infrastructure of fully automated driving on the highway (Figure 2). However, the requirements for the range of the sensor system are not as high due to the considerably lower driving speeds on a parking lot. Instead, precision down to the last centimeter is required for self-localization and manoeuvring. The technical implementation is by far the most advanced today for automated parking. Initial series applications have already been implemented and, due to the stable starting point, this will very quickly evolve to valet parking in dedicated areas.
Automation, Step 3: Reaching Destination WithRobo Taxi
Now the driver and passengers are precisely where they need to be to continue the last part of their journey seamlessly, without any walking at all. An automatically pre-ordered self-driving vehicle drives up to the mobility terminal (Figure 4). There are many names for this new category of driverless vehicle in accordance with SAE/VDA Level 5 driverless, such as “People Mover”. We will call it a robo taxi to make things simpler. This fully automated vehicle transports travelers right up to the door of their destination, which spares them any difficult navigation through a big city as well as the hunt for a parking space.
Compared to your personal car with the optional automated driving function, the robo taxi no longer has any operating elements for a driver. It is therefore a vehicle without steering wheel or pedals. It is used in inner city traffic at comparably low speeds but a high level of environmental complexity.
Continental will set up a shuttle service with the innovative test platform CUbE (Continental Urban MobilityExperience) on its company premises in Frankfurt this year to test the practicality of robo taxis. The electrically-drivenrobo taxi connects the main building with the onsite test track.
There is another difference between the requirements for the automation of a robo taxi and those of a private passenger vehicle. Studies have shown that a typical private passenger vehicle is driven less than one hour a day on average. This means that, with the exception of longer journeys, these vehicles are stationary 23 hours a day, or are not used at all. The situation is quite the opposite for a robo taxi. The profitability of this vehicle class depends on the robo taxi driving as much as possible and generating revenue, just like in the logistics sector. Modification of the operating models and use of electricdrives changes the requirements profile of the brake components as well. The thermal load created by deceleration from high speeds is an important design criterion for a brake system today. In comparison, the number of actuations during daily use is the main component taken into consideration for a robo taxi. These changes in components are being examined at Continental and taken into consideration as they are developed for future mobility concepts.
The conviction is growing in many parts of the automotive industry that robo taxis will play an important part in urban mobility in the future. So it hardly comes as a great surprise that a large number of companies and research institutes are tackling the development of this new type of individual mobility.
Today’s form of individual mobility presents problems in many parts of the world. Singapore and Japan are prime examples of this. The particular combination of enormous congested urban areas and very sparsely populated rural areas is problematic. For this reason the Japanese government has announced the widespread use of autonomous vehicles during the Olympic Games as a measure for coping with the numerous visitors. Even today the government is very actively encouraging technological developments in this direction. This is an excellent opportunity for Continental to develop technical solutions together with public administrations in actual areas of application.
Thanks to different developments in automated driving, it is possible for seamless transition between individual sections of a journey and different transport vehicles. The requirements of the AD solution and participating components change according to the section of the journey. From the current perspective, automated driving level SAE/VDA 5 (driverless) will be the next step in supporting seamless mobility with free flowing transitions between the most comfortable modes of transportation on a journey.
The range and complexity of the participating technical systems is great and can only be provided by a technological company with a very broad base and an integrated approach. Continental is driving the development of all types of AD forward so that, in future, we can offer solutions to everyone for challenges such as those presented by the planned use of robo taxis during the 2020 Olympic Games in Tokyo.
Automated driving places high demands on the innovative capacity of the big players in the global automotive supplier industry. At Continental, we believe it is our responsibility to make intelligent and sustainable mobility possible and to continue to develop its efficiency to achieve Vision Zero in the long run, the vision of accident-free driving. Across the group, Continental is working with well-known colleges and universities to research the potential of artificial intelligence for decision chains in automated driving