Keywords

1 Introduction

To enable all passengers, including those with special requirements, to benefit from the new public transportation services introduced by H2020 AVENUE, we need to fully understand their requirements and expectations, their needs, the issues and the problems, and also the personal tricks and strategies they have developed for themselves. Will they accept a driverless bus? What if there is no real person on board? What does the absence of a driver or conductor imply for older persons with disabilities and persons with reduced mobility (PRM) (European Commission, 2014)? What are passengers’ expectations in their interaction with the vehicle?

2 Requirements of Passengers (Interview Results)

To gain a real understanding of the user, the AVENUE project used a variety of methods to acquire the needed information. Since people with disabilities have special requirements, they play an important role in gathering user requirements.

Disability is part of the human condition. Almost everyone will be temporarily or permanently impaired at some point in life, and those who survive to old age will experience increasing difficulties in functioning. (World Health Organization, 2011)

Considering that every human will have some type of disability in his/her life or can be situationally induced disabled (e.g. wearing a headset), the requirements derived from people with disabilities will be useful for all (Fig. 8.1).

Fig. 8.1
A table with 7 columns and 15 rows. It gives the respective age of deterioration for 7 elements related to vision, 3 related to hearing, and 1 related to touch. It includes increased need of light from 35 years onward and reduced hearing ability in men from 32 years onwards.

Deterioration of abilities in the process of ageing based on Saup (1993) (Translation by the author)

In a first step in 2018, interviews with public transport users were conducted in five European states. This survey had its major focus on public transport in general, as at this point in time hardly anyone has had real experiences with AVs.

In 2019 and 2020, when AVs had been in operation at several sites for a while, we were able to gather real passenger experiences with the AVs in six European cities. We conducted interviews with passengers as well as with safety operators in the busses, and we observed passengers in the busses and at the stops. Another information source were workshops with mobility experts (with and without special requirements) and public transport operators.

2.1 Phase 1 (July–September 2018)

In phase 1 (July–September 2018), together with the local AVENUE partners, we conducted interviews with public transport users in five countries: Denmark, France, Germany, Greece, and Luxembourg.

In this phase, we had our focus on public transport in general. Our goal was to fully understand the needs, the issues, and the problems of public transport users and the personal tricks and strategies they have developed for themselves.

The interview was divided into three parts:

  • Requirements for public transport due to the well-known use of “classic” public transport.

  • The aim of this part was to identify the status-quo and current issues.

  • Evaluation of experience with fully automated vehicles (if existent), especially in comparison to conventional public transport.

  • Wishes and expectations for future fully automated vehicles.

  • In this part, all passengers were asked to imagine how future fully automated busses could change and improve public transport.

A total of 58 persons participated in the interviews, and 33 of them had some kind of disability, most of whom were visually impaired.

2.2 Public Transport in General

Most of the results of the surveys on public transport in general are as expected:

Passengers dislike delays, low frequency of service (and thus long waits), crowded vehicles/no seats, dirty and littered vehicles, lack of information, and unfriendly staff. They want on-time public transportation, affordable tickets, increased frequency and destinations, clean vehicles, and available and comfortable seating, including guaranteed seating for people with special needs. Accurate and accessible information, e.g. audibly understandable announcements, accurate (not delayed) announcements of next stops, estimated time of arrival, and current location of the bus are always requested.

It would be nice if the stops were more flexible: For some participants, it would be a real advantage to be picked up from home and dropped off at their actual destination, rather than at a nearby (or not so nearby) bus stop. This way, they would no longer have to walk all the way to a bus stop and could live a more mobile and independent life.

2.3 Attitude Towards Fully Automated Public Transport (Unexperienced Pax)

Most interviewees in 2018 have never used or even seen an AV. An evaluation of their experience with AVs was thus not possible. Nevertheless, the interviewees were asked about their attitude towards fully automated vehicles. Many interview partners are quite sceptical and are not convinced that the technology is already mature enough to be trusted. They consider traffic situations too complex to be handled by technology.

Some even say that they would never use a fully automated bus (7 persons out of 58).

A majority needs to gain more trust in this technology before they are ready to get in. Only four interviewees are ready to hop on a fully automated bus at the day of the interview.

But it turns out that fully automated public transport is accepted more easily with increasing distance to the next stop (Fig. 8.2) and with decreasing frequency of public transport.

Fig. 8.2
A triple bar graph plots value versus distance to closest P T stop. The data is for yes, no, and under some circumstances. The bar for yes is the highest for greater-than 500 meters at 0.83. The bar for no is the highest for less than or equal to 100 meters at 0.46. The values are approximated.

“Would you take a driverless bus?” in relation to the distance to the next bus stop

Unexpectedly age seems to play no role (Fig. 8.3).

Fig. 8.3
A double bar graph plots number of answers versus age group. The bars are for yes and no. The highest bar for yes is for greater than 60 at 15 with no at 9. For ages less than 24 yes is at 1 and no for 2. The values are approximated.

“Would you take a driverless bus?” in relation to age group

Safety is an important topic for the interviewees. Most of them have heard of accidents by fully automated vehicles, and they stress that this technology needs extensive testing before it can be put into operation. They would use fully automated vehicles only if they are convinced that they are safe.

While they think that technology like sensors, etc. provides advantages, many users nevertheless fear accidents. Some believe that other drivers will crash into the bus as they will not be able to anticipate its behaviour. Others believe that the technology is not fail-safe and/or advanced enough to handle complex traffic situations, and the fully automated bus will have accidents without a driver. Only a minority of interviewees is of the opinion that safety will be increased.

Many interview partners are afraid that the use of fully automated busses in the field will lead to more delays and failures due to unstable technology and because traffic situations are too complex to be handled by technology in general.

Some are even worried about cyberattacks against a fully automated vehicle: if there is no driver to interfere, hackers could make the bus go faster or drive off a bridge or into oncoming traffic.

The overall attitude/scepticism towards fully automated vehicles makes clear to us that trust has to be gained first, and even little incidents or accidents are likely to destroy that trust.

There is a great empathy for drivers who might lose their jobs. Most respondents cannot imagine fully automated vehicles without having this in mind. This is especially reflected in the fact that when talking about positive experiences in public transport, respondents include other people, especially the bus driver (receiving help, amusing passengers, a funny bus driver, etc.) and in the fact that 60% of participants mention talking to the driver (half of them for a conversation).

For most interviewees, a “safety operator” who can interfere or take over and acts as an authority figure is essential. Passengers also want someone in the bus who can answer questions, provide information, and help them getting on or off the bus when necessary. They are afraid that vandalism or even robberies or assaults could be a problem if there is no supervisor in the bus.

Besides all the scepticism, the interview partners also have some expectations that reflect the above-mentioned wishes for public transport: They assume that there will be connections where there are none today, because they are not profitable. Many also expect a smoother ride by the machine than by a real driver and cheaper tickets and that busses will not have to be cancelled due to lack of available drivers.

2.4 Phase 2 (June 2019–February 2020)

While our focus in 2018 was public transport in general and anticipated experiences with AVs (due to lack of real experiences), 1 year later our aim was to go beyond anticipated experiences and to figure out the issues, the problems, and the missing services in fully automated vehicles. Therefore, all participants in this second series of studies have already gathered some experiences with AVs. To get a wide range of information, we decided to follow several different approaches to gather information: interviews with safety operators, interviews with (disabled) passengers, and observation studies.

2.5 Interviews with Safety Operators

Currently, every fully automated vehicle in the AVENUE test sites (and the visited non-AVENUE sites) has a safety operator on board. These operators are responsible for the safety onboard and can control the bus manually if necessary. Furthermore, they are contact persons for the passengers. The safety operators are a valuable source of information as they have direct contact to passengers and experience their reactions, open questions, requirements, and problems every day. There is probably no other person who has had more user contact and knows the questions and problems of the users better.

In order to obtain the most information from this valuable source, we have decided to give questionnaires to the safety operators of all participating PT providers and additionally conducted some interviews with safety operators in non-AVENUE sites like Berlin, Bad Birnbach, and Vienna.

A total of 27 safety operators (5 from non-AVENUE sides) have been interviewed and asked about their impression and experiences made with passengers onboard of the vehicle. All of them mention a positive acceptance of the vehicles by nearly all passengers. Most users (especially the anxious ones) talk with the safety operators about the technology. It is often the first topic of conversation. Questions like how the bus knows where it is and how it works are dominating here. The better understanding of the technology helps them to relax and enjoy the ride. One operator mentions that even those passengers with a negative view change their mind while driving with the vehicle.

Nearly all passengers like the vehicle, and also the arrangement of the seats is welcomed because it encourages communication. However, all are complaining about the hard and unforeseeable braking. So smoother braking is one of the most wanted features, besides more speed.

Typical situations in which the drivers help the passengers are getting in and out (including opening and closing the doors) and buckling up. The high boarding height of the minibus is a real disadvantage and frequently requires the support of the safety operator. Some drivers report that they have to calm passengers after one of the above-mentioned hard brakes.

The question about the operators’ needs and wishes results in interesting answers and ideas, and the most popular answers and suggestions are:

  • Colour coding of the seat belts (which are currently not marked clearly)

  • A smoother braking behaviour (has been solved already)

  • A stand or a seat with seatbelt for the safety operator (back pain is already an issue)

2.6 Interviews with Experienced Passengers: Attitude Towards Fully Automated Public Transport

In our interviews in 2019, we only asked persons that have already gathered some experiences with AVs, i.e. they have made the decision to use an AV. So it is not surprising that the answers are much more positive and less sceptical.

We interviewed 18 passengers in Berlin and Vienna, 7 of them with reduced mobility.

The interviewed passengers are very interested in the shuttle and its technology. They want to know things like What can the shuttle “see”? Why and when does the operator drive?

Many passengers complain about the low speed and that the vehicle comes to a stop too often. In order to be a useful means of transport, it needs to go faster. Everybody wants a smoother braking behaviour—as mentioned above, this issue seems to be solved by now.

A huge topic is information. Our interview partners miss a real passenger information system (PIS). The monitor in the shuttle is too small, its position not ideal (cannot be seen from every seat), and combined with the used font size and contrast, it is not considered usable.

Some passengers complain about the missing announcement of the next stop.

As there have been many complaints about missing information, we asked users more precisely what kind of information they would expect. They clearly state that they want at least the classical information one gets in standard PT as well as some information specifically required in this new type of transport:

  • Announcement and display of stops (next stop/current stop)

  • Transfer options in combination with time to transfer

  • Information on how to request a stop

  • A warning that the vehicle is about to brake

  • Info on shuttle’s battery: battery status and required charging time

The request for information on the vehicle’s battery might seem unexpected at first. One reason for this could be that some passengers have experienced that the service had to be stopped due to an empty battery caused by the heavy usage of the air condition. Another reason could be that e-mobility is becoming more common, and people are therefore simply curious about such details.

Our interview partners also ask for information at the bus stop or from outside the bus:

  • Information (at the bus stop) when next shuttle will arrive

  • Number of free seats in the bus

The request for this information might be due to the fact that the vehicles are rather small and therefore the number of seats are limited.

The passengers appreciate the presence of a safety operator on board the AV. They think it is good to have someone in the shuttle who can intervene in an emergency and who can help to get on the bus with a stroller, etc.

Overall, passengers enjoy the experience, they trust the technology, and they would use the shuttle again at any time and recommend it to friends.

Most interviewed passengers say that they would use the shuttle.

The PRMs are even more willing to do so as they rely on public transport and are willing to take advantage of everything that makes life easier. All passengers agree that they would consider using the fully automated shuttle on a regular basis if

  • The shuttle went faster

  • The routes were longer

  • There was a regular service with frequent departure times

  • The shuttle could deal with everyday traffic situations

  • There was an on-demand service with flexible routes (including bus stop in the immediate vicinity of the place of residence)

  • Information was available at bus stop when next shuttle will arrive

  • Infrastructure was improved: coordination with other means of transport like subway

  • There was more space for strollers, etc. (currently only space for one stroller, buggy, etc.)

Many users like the idea of a shuttle service that does not depend on fixed bus stops. However, they raise the topic that the shuttle may stop only at secure places. It is not clear how the passenger could communicate these places to the shuttle and how the shuttle could confirm this. Other interviewees are concerned that the bus could not keep to the schedule without fixed stops. Some people could make the bus stop all the time just for fun.

Of the 18 interview partners, 11 instantly say they would use the fully automated shuttle even if there was not a safety operator on board. They would like to have a hotline available in the shuttle so that passengers can call if they feel the need to do so.

Some interviewees hesitate and would drive with the shuttle without safety operator only if the shuttle was more reliable and did not constantly stop whenever there is an unexpected situation like an illegally parked car.

Four interviewees are firmly convinced that the shuttle without an operator is not possible at the present time as this was too dangerous in road traffic. Some think that a human attendant will always be required in the shuttle. What if the bus suddenly stops and does not drive on? What if someone gets stuck in the door? What if there is an accident?

3 A Blind Users’ Perspective on Automated Vehicles

To illustrate the challenges for PRMs, we accompany our persona Klaus-Dieter on his first trip alone with an automated e-minibus.

Klaus-Dieter, a 59-year-old male person, has good experiences in terms of mobility. In his hometown, a medium-sized city in Germany, he daily uses public transport completely on his own, from home to work and vice versa.

3.1 Bus Stops on the Course

The automated bus, which is tested by Klaus-Dieter, runs in a kind of circular course and provides ten different bus stops where the vehicle automatically stops to allow the passengers to get in or out.

Regarding Klaus-Dieter, the test environment is completely unknown for him, and therefore he is not able to find his first bus stop on his own without an assistance, even though the address of the corresponding bus stop was given to him in the lead-up to his test.

All bus stops are represented by normal traffic signs with additionally installed timetables, directly mounted at eye level on the pillars of the traffic signs. The bus stops are neither equipped with audio signals (e.g. attention signals on traffic lights) nor with tactile markers on the walkway, which could help blind users to find and quickly locate specific places on their own.

The next difficulty occurs while Klaus-Dieter tries to get the arrival and departure times from the timetable. The table is not provided in braille, and an audio-based passenger info system, e.g. a standard loud speaker, is not available. Nevertheless, online-provided timetables as well as the usage of personal mobile phones together with appropriate OCR apps can solve these kinds of problems in the future.

The automated bus is not equipped with AVAS; therefore the arrival of the bus is difficult to detect for Klaus-Dieter, who is not able to perceive road noise from the bus due to environmental sounds.

3.2 Boarding Process

The bus directly stops at the bus stop sign but without any attention signal. The door does not open automatically, and Klaus-Dieter has to locate the corresponding sensor button that is mounted on one site of the door of the bus but without any kind of tactile markers. Moreover, there are two different sensor buttons available, the second one is for wheelchair users who can request a special ramp for boarding.

First challenge, which of the buttons is the right one? There is neither braille nor tactile markers available to answer this question in an accessible way. Audio signals which could help blind users to quickly locate the door as well as to find the appropriate sensor button are also absent, and Klaus-Dieter has to cope with this issue for himself by trial and error.

There is no audio-based passenger information system installed at the bus stop which could enable Klaus-Dieter to identify the arriving bus as his correct line.

3.3 Interior Situation and Bus Ride

All passengers of the bus are assigned to take their seats during the bus ride to reduce any potential transport risks. Standing passengers are not allowed. Klaus-Dieter assesses the number of grips, handles, and handrails to be too low to ensure a safe movement or orientation within the interior of the bus. He also has problems to quickly find his seat without assistance. He was disappointed, as there is no additional storage space for hand luggage, shopping bags, or guide dogs. The riding process is comfortable, there are no abrupt braking actions at all, and all speed-up or slow-down actions lead Klaus-Dieter to a good driving experience in terms of security requirements.

The automated bus is configured to operate in metro mode; this means that the bus automatically stops at all available bus stops at the circular course. There is no audio-based passenger information system installed which could tell blind users the next bus stop in an accessible manner. Therefore, Klaus-Dieter has to know the stops at the circular course and moreover to count the already visited bus stops to get out at the right one.

The security driver explains Klaus-Dieter that there are no complex traffic situations within the bus course, e.g. complex set of traffic lights or difficult crossings. Potential stops of the automated bus are only caused by two situations: the vehicle reaches the bus stop or the vehicle detects obstacles on the course. In both cases the bus indicates the stop by a clear and good perceivable audio signal but without any audio information of the corresponding reason. Arrival points are not announced by speech. Moreover, it is the same signal for both situations, and Klaus-Dieter is not able to differentiate between the above-mentioned triggering events. As a result, he tries several times to get out of the bus on the course but without success.

3.4 Getting Out of the Bus

In metro mode, the bus stops at every stop, but the doors have to be opened manually, in terms of accessibility blind users are faced with several challenges.

Despite the already mentioned missing acoustical information, the blind user needs to figure out how to open the door. Klaus-Dieter remarks that the placement of the different sensor buttons, each one equipped with a specific function, will lead blind persons to operating errors. These difficulties are caused by the sensitivity of the sensor buttons that instantaneously trigger in case of any finger contact, but finger contact is required for this user group to read braille as well as to identify other tactile markings.

Due to these problems, Klaus-Dieter is unable to distinguish between the three halt functions, provided by the different sensor buttons: SOS, halt request for wheelchair users (which automatically activates a ramp), and the default exit button that opens the door. In particular, the exit button raises additional problems because the door must be released by the automated bus security system before passengers are able to successfully trigger the exit button to leave the bus. And exactly this release feedback is not perceivable by Klaus-Dieter who is only able to get out of the bus after pressing several times the different buttons to open the door.

3.5 Klaus-Dieter’s Summary

Some of the above-mentioned accessibility issues can be resolved by using the capabilities of a modern mobile phone, e.g. GPS localisation, and online timetables could help users to successfully deal with these specific problems. But regarding the described topics, there are a lot of additional improvements required at the busses to answer the needs of this user group and to ensure a comfortable public transport with automated vehicles.

4 Situation-Based Impairments of Different Passenger Groups

Additional mobile and audio information is required for the following user groups on the go, at the bus stop as well as in the bus:

  • International tourists or other passengers who are not familiar with the local language.

  • People with dyslexia.

  • Persons with misted-up glasses, particularly in winter times or with an FFP2 mask.

  • Younger children who better cope with audio than with visible feedback; audio feedback grants a higher real-time information awareness for this specific passenger group.

  • All persons who use their mobile phones to read articles (or use the Internet) during the bus ride; appropriate audio info raise their attention to get the next bus stop just in time.

  • Smaller groups of passengers who intensively discuss with each other often get lost within their conversations and therefore miss the exit at their final destinations; appropriate audio feedback can avoid such situations.

  • Persons with hearing impairments need an appropriate kind of suitable feedback in all above-mentioned situations, e.g. vibration or other haptic feedback, to raise their awareness in corresponding situations.

5 Proposed Implementation of User Requirements

The project co-created a mock-up for an accessible app with disabled users for fully automated public transport. Furthermore, a proposal for an in-vehicle information display to meet the user requirements in a fully automated vehicle was created.

5.1 Mock-Up for an Accessible App for Fully Automated Public Transport

Based on the challenges Klaus-Dieter has experienced, the project co-created with disabled users a mock-up for an accessible app that supports all users to use fully automated public transport (Fig. 8.4).

Fig. 8.4
A screenshot of a mock-up for an accessible app. The text at the top reads upcoming arrivals. It is followed by 4 tabs with trains arriving on different times with available seat numbers.

Mock-up for an accessible app

The app covers the complete passenger workflow:

  • The way to the bus stop

  • Waiting at the bus stop

  • Identifying the correct bus

  • Entering the bus

  • Riding on the bus

  • Leaving the bus at the desired destination

5.2 Information Display in the Vehicle

As most interviewed passengers ask for a passenger information system (PIS) that goes beyond the classical information we know from public transport today, we developed a concept for displays in the fully automated e-minibus (Fig. 8.5).

Fig. 8.5
An illustration of the wireframe of information display. Classical P I S, 1, is at the top left. Sensor view, 2, is at the top right. Map view, 3, is at the bottom.

Wireframe of information display

It consists of three major components:

  1. 1.

    Classical PIS

  2. 2.

    Sensor view

  3. 3.

    Map view

The passenger information system provides a list of upcoming stops including information on actual (not planned) arrival times and delays. If there are any incidents or restrictions on the route, this is also displayed here. The number and destination of the bus are shown as well.

The sensor view allows the passenger to “see through the eyes of the system”. Stewards report that many passengers ask questions about the capabilities of the vehicle, especially regarding the sensors, so by indicating the sensor range and detected objects, we can satisfy their curiosity and at the same time build trust in the technology.

The map view takes up the most monitor space. It answers the questions “Where am I right now?”, “Where am I going?”, and “What is nearby?”. This especially benefits passengers who are not that familiar with the area. For them, the name of the bus stop is often not meaningful. The map with marked landmarks can support their orientation in the area and help them find the best way to their destination and the stop where they should leave the bus.

In addition, sights, restaurants, and stores along the route can also be displayed on the map so passengers can see what they can reach from the bus (Fig. 8.6).

Fig. 8.6
A screenshot of the suggested information display in the vehicle. It has time, speed, coast line, and a map with stops.

A suggested information display in the vehicle

6 Conclusions

The user acceptance increases to the same extent as the fully automated vehicles are optimised and adapted to fit the different user requirements.

Passengers with and without disabilities differ in their needs, but both user groups expect a significant added value in the use of automated minibuses for public transport, especially in areas with poor public transport coverage.

While the problems of people with (temporary) physical or sensory impairments are more related to the practical use of the technology, e.g. finding the right bus, finding and activating the door used, and missing PIS, other passengers tend to see difficulties in buying their tickets and in connecting to other public transport systems.

In order to make automated public transport a solution for all, all the services provided by the driver must be provided in some other way according to the dual-channel principle.

This also includes the perceived safety and security.

In addition, low-level access is essential, especially for passengers with walkers or buggies.

However, the prerequisite for use of automated vehicles is a general trust in the technology.