Automated Vehicles Bill [HL]

Baroness Bowles of Berkhamsted Excerpts

Committee stage
Wednesday 10th January 2024 Lords Chamber

Moved by

[Baroness Bowles of Berkhamsted]

1: Clause 1, page 1, line 14, at end insert—
“(3A) Vehicle testing must include substantial real testing on roads in the United Kingdom in addition to simulation testing.”Member's explanatory statement
This amendment would probe the intention with regard to real as well as simulation testing for UK road situations.

[Baroness Bowles of Berkhamsted (LD)]

My Lords, I think that this is the first time I have had an Amendment 1, but, in any event, it gives me pleasure to start the Committee stage of the Bill. For the purposes of Committee, I declare my interest, in that a family member works in the vehicle connectivity sector, but I have no financial interests.

I have three amendments in this group: Amendment 1, and Amendments 20 and 27, which are the same text appearing in different clauses. Amendment 1 is very much a “does what it says on the tin” amendment, and states that vehicle testing must include substantial real testing on roads as well as simulation testing for UK road situations. As well as for initial licensing, this may also have relevance when vehicles licensed in other countries are brought here, especially when driving on different sides of the road and road signs are differently placed. I was prompted to put in this rather obvious statement because among the various things that I read in the documents it was pointed out that simulation testing for UK road situations would be allowed—and I can accept its usefulness as an element when converting from well-proven automation on roads in other countries, for example. However, what I cannot accept is simulation on its own being sufficient, and I wish to ensure that that is not the case.

A further reason for this amendment is that I am aware of how, in the US, there have been issues moving from one city location to another, because of different road widths, despite those having been simulated. Noble Lords who do transport all the time can probably identify what I have read, but I am sure that moving from Los Angeles to the UK would have even more issues, including, for example, more narrow, ancient, humpback or bendy traffic bridges without traffic lights where it is possible only to go one way at a time.

Despite having come up with amendments, I take the approach across this legislation that I understand it is an enabling framework and will not contain detail and, further, that with consultations and so on, a broadly sensible approach will result. Nevertheless, when we have been given documents that explain current thinking and direction, they also explain that they are not fixed promises—presumably because there is still quite a lot of work to do and we do not yet know what the priorities will be. From looking at other amendments generally, it seems that other noble Lords also think we need a few more fixed promises on things that we can be certain will not be left out, and therefore seek to have them in the Bill. For me, real UK road testing, rather than only simulations, is one of them. Obviously, within that, I would expect the road testing to apply to the roads on which the vehicles will be licensed for automated use: on motorways for motorway driving, in towns for town driving, and country lanes with single-lane passing places—if you are lucky—for country lane driving. Will the Minister confirm that this will be the approach, and can we have assurance by some text in the Bill?

My other two amendments, Amendments 20 and 27, relate to adding insurance and captive insurance into the provisions that establish the financial soundness of an authorised self-driving entity. The Law Commission referenced insurance as being able to provide part of the financial soundness, and I would like to see that included, rather than it being thought an additional measure on top of everything else.

I also raised the issue of captives with the Minister at Second Reading, and I thank him for his reply. In the Bill, I would like to see captives acknowledged alongside mainstream insurance as an acceptable form of insurance in the context of ASDE financial stability. Call me cynical or pedantic, or probably both, but I have had too much involvement in financial services and insurance not to think that it needs specific elaboration to ensure that captives, as well as independent insurance, can be considered as an element of the financial stability package.

As I said, I found insurance mentioned in the Law Commission documents as a possible part of the financial stability assurance, so can the Minister say whether there was any specific reason for not following suit and not mentioning it in the Bill? If there were no specific reasons, will the Minister be inclined to recognise my warning, as there might be quibbling if it is not specified? I beg to move.

[Lord Lucas (Con)]

My Lords, I have a number of amendments in this group. I begin by asking my noble friend the Minister to encourage his team to get me a reply to what I call my Eastbourne email. I hope to use that as a means of understanding exactly where the Government find themselves with a practical example of an early-stage project. It would be helpful to have that by the second day in Committee. If I have already received it, I have missed it, so I would be grateful to him for pointing that out to me.

What binds the amendments in this group together is, first, that I do not expect these things to appear in the Bill because I think that they are covered. But they are covered in a way that does not make it clear what the Government will actually do, so I hope to draw out of them some information on what their intentions are.

Secondly, the amendments encourage the Government to take the standards-setting process seriously. I have some long experience observing the telecommunications industry. That has faded from my early days in the City, when we were one of the dominant world players, to now, when we are nothing. Part of that decay has been because we let standards-setting slip. If you want to be a place where a new technology is establishing itself and where companies want to come and be part of what you are doing, being part of the standards-setting is absolutely key. You have to assign good people to it—people who will be internationally respected for their views and insights in the industry—and give them the time to make a really serious contribution to the process. It is then independent of what is happening in the UK; they become part of the wavefront of what is happening, because the whole standards-setting process involves understanding the way things are going, what is happening and who is doing what. That information then flows back into the structures in the UK, and you get a local understanding of where the opportunities are and how the UK might take advantage of them.

If we had had that with telecommunications, we would not be in the dire state we are in now. We started with huge advantages, but they have all gone. Here we are with a new industry and a very clear need for international standards, so we absolutely must take that seriously and put our backs into being part of that process.

I will pick up on the individual amendments. The vehicle identification system—the way in which vehicles will say, “Hi, this is me”—will clearly be electronic. The whole business of using number plates has broken down, and there are 10 million or so unauthorised vehicles on British roads, for all sorts of reasons—vehicles that are just not known to the DVLA, are not taxed and have strings of outstanding parking tickets. Nobody knows whether a number plate they see is real or cloned. We do not need this happening in a new industry, where it will be really important to establish exactly which vehicle was doing what and at which time. It has to be an electronic system, it has to be something that is embedded in hardware, and it absolutely has to be consistent internationally. A vehicle coming over from the continent has to use the same system. This is an example of something that we have to develop and a direction we have to go in, and we absolutely have to be part of setting that standard.

Amendment 15 looks at the question of a passenger alarm. If you are in a vehicle that is travelling totally autonomously and something is wrong and you want to raise the alarm, how do you do it? What is the system? What should you expect to find in the vehicle? Are we going to restrict travelling to people who happen to have mobile phones on them at the time? I hope not. What is the system to be? Again, we ought to be part of establishing international standards, because we want to be able to admit vehicles to the UK. This should be about not just our own domestic expectations; there should be something running internationally.

We want vehicles to be able to communicate where they are and, if they are part of some kind of lending, taxi or other scheme, whether they are available. Again, this needs to be done in a standard way, so that different owners and manufacturers are all sending this information out in a consistent way, and on the back of it can be built the sort of systems consumers will need to know whether or not an autonomous vehicle is available to them. We should not reach a block or allow this to become balkanised, with different companies owning little bits; the information available to consumers ought to be clearly available to everybody.

Amendment 17 looks at the process of reporting on the condition of vehicles, as there are various bits of the Bill that make it clear that automated vehicles are expected to be well maintained. If a vehicle detects that it is not in the state that it ought to be in, that needs to be reported. It needs to be reported not just internally to the system but in a way that makes that information, and the fact that it was reported, available to investigating authorities. Again, we need a standard for that, and it needs to be an international one.

Amendment 18 looks at the question of waymarkers: how a vehicle knows exactly where it is in a relatively autonomous landscape. Are we going to be totally reliant on the navigation satellites working or are we going to have a more ground-based reference system? Some manufacturers clearly think that they will have within their vehicles an image of the routes that they are taking and that the vehicles will recognise where they are. That is a darned hard thing to do on some motorways—you just do not know which bit you are on, or indeed which country you are in: “Am I in Germany or am I still in the UK?” There is a system on motorways where, in the physical sense, you can look at the waymarkers—if you are not travelling too fast—and see where you are; if you break down, it allows you to read the sign and say what distance from it you are. Are we thinking of building that into automated vehicles?

Lastly, how will vehicles communicate with the emergency services, whether it is a fire engine coming up from behind asking the vehicle to pull over and let it through or a policeman standing at the edge of the road, waving down the vehicle to stop? How will that be achieved? Again, we will want there to be an international standard; we do not want to find that vehicles coming in from abroad are unable to speak English. There has to be a common system in there somewhere. However, we absolutely want it to happen—we do not want our police to be powerless and for the automated vehicles to sail past them because they do not understand a hand wave. There has to be some communication system. There are lots of options, but we have to specify it.

--- Later in debate ---

[Lord Davies of Gower (Con)]

I am sorry if the noble Lord took that view of it, but that was not my intention.

[Baroness Bowles of Berkhamsted (LD)]

My Lords, I thank the Minister for his responses. At this stage, of course, everything is probing. I possibly still entertain a hope that we can have some little light-touch mentions that are not overbearing somewhere in the text. Maybe we will return to some of these issues on Report.

There are one or two things other noble Lords have said that I would like to touch on. The noble Lord, Lord Lucas, mentioned connectedness. We are falling into a bit of a trap if we start talking about the connectedness of automated vehicles, because the big prize is the connectedness of all vehicles—those which are driven and those which are automated. That is where the real benefits to traffic management and the economic benefits reside. That is a much bigger scheme of interconnectedness, and we are doing ourselves a disfavour by almost sidelining the connectedness and connected car issues as if they are something small and of less importance than the big goal of automated vehicles. In the near term, connectedness is a lot more relevant and moves into what is happening with automated vehicles. We should try to think of it as more of a whole.

I am aware on the simulation aspects, which were addressed, and that we cannot have millions of hours of road driving. Simulations are important and it is an iterative process between simulated tests and road tests. I am perhaps reassured that that is what is in mind. I still do not like the vision that, sometime in the future, it might happen that there are absolutely no road tests—even small ones. Maybe it is wrong to try to insert “substantial”, implying that—this is not what I intended—it would be more than the simulated tests. I still think there should be a significant amount in there for a very long time into the future.

The noble Lord, Lord Liddle, said that his main interest is safety. Certainty is quite fundamental to safety. There is lots more to get to, so I will not say any more now. With the notion that I might return with this in a gentler form on Report, I beg leave to withdraw my amendment.

Amendment 1 withdrawn.

--- Later in debate ---

When we come to Report we will almost certainly pursue these two points—the careful and competent driver as the test, and having affirmative secondary legislation so we can be fully involved.

[Baroness Bowles of Berkhamsted (LD)]

My Lords, I have two amendments in this group, which add to the safety principles. Amendment 7 would add “road environs” to Clause 2(2) so it reads that

“The principles must be framed with a view to securing that road”


and road environs

“safety in Great Britain will be better”.

I had two broad points in mind—one that it is relevant what happens next to the road, on pavements, driveways or anywhere else that a vehicle might stray if navigation goes wrong. It would be relevant, for example, if a consequence of some event meant that the vehicle swerved off road instead of stopping. The swerve might be safer for the road, but the vehicle might hit people not on the road, so it would not be safer for the road environs. I accept that the general standard is to stop not swerve, but that was an easy example to give. It is an obvious point, but something relating to the environs needs inclusion and the statistics that are analysed need to take those kinds of things into account.

I happened to come across a paper today—it was actually published yesterday—entitled Unreliable Pedestrian Detection and Driver Alerting in Intelligent Vehicles, by Professor Mary L Cummings, a senior member of the IEEE and a professor of mechanical engineering, electrical and computer engineering and computer science at George Mason University, and Ben Bauchwitz from Duke University. They have done some testing to try to detect pedestrians and, as the title might indicate, it did not work out all that well. Among the suggestions are that

“intelligent vehicles … detected the pedestrian earlier if there were no established lane lines, suggesting that in well-marked areas, typically the case for established crossings, pedestrians may be at increased risk”

because of the road markings. Obviously, these are all kinds of things that we have to take into account: it just shows that we have to look at what is happening in the whole environment.

There are other things that are going on in, around and among roads that are not part of whatever connected systems are developed, whether it is pedestrians, cyclists, animals that can be ridden or animals in the wild. Of course, we have plenty of such roads, where sheep graze in the Dales and ponies in the New Forest: they are not going to be part of the connected systems, so we need to be sure that the actions of those are taken into account. Less picturesque than those but omnipresent—I flagged them in my reasons—are delivery vehicles. Delivery vehicles already have a big and frequently annoying effect on roads. I doubt that I am the only person who has experienced near misses caused by bad or inconsiderate driving, or an inability to see the road ahead due to dangerous stopping by delivery vehicles, and there is no doubt that the tight scheduling of drivers bears some of the blame for that. Of course, we are hoping that automated driving will be more observant of legalities, but several noble Lords mentioned delivery vehicles at Second Reading. There are papers that explain how little robots are going to be coming up your drive, so what is the situation there? What testing will there be with delivery vehicles that are going to be partly on the road and partly going into private driveways?

An interesting point here is that, when I submitted my amendment, my explanatory statement had to be truncated to remove reference to private driveways because that was out of scope. It seems to me that the Bill is only about public highways, but we cannot get away from the fact that private driveways and private roads are pretty abundant, so what is the legal situation there going to be? Because that is out of scope, is it abandoned? Presumably, regulations cannot be being made, and I cannot help feeling that this is a little bit of a hole. The closest I could get to it was by “road and road environs”, which at least seemed to pass the sniff test in the Bill Office. Thus, in connection with both these amendments, my question to the Minister is: how much will testing and licensing take account of effects that are beyond the highway? What is actually included within the “highway” definition, so far as the Bill is concerned, and what is left out?

I have quite a lot of interaction with the highway, because I live with one going all the way up alongside me, and it is quite remarkable, from time to time, what the local authority thinks is part of the highway but is actually a 130 year-old ancient hedge that they wanted to chop down. Anyway, the corollary to all this is that, if testing and authorisation is done only in the context of highways and what happens there, what is the legal framework for these private and residential roads and driveways? If they are left out, are we going to have something in addition?

[Lord Berkeley (Lab)]

My Lords, this is a very interesting and incredibly important group of amendments. My noble friend Lord Tunnicliffe’s introduction was masterful in setting out all the problems. Before I comment on them, however, I would like to comment on a remark by the noble Baroness, Lady Bowles, just now about which bits of the country, whether they are highways or not, are covered by this legislation. A few years ago, it was nothing personal but I had to investigate whether somebody who was driving a vehicle under the influence of alcohol on an unmade road—in other words, a private road—could be guilty of drink-driving offences. The answer was that they would not be guilty of just about anything apart from drink-driving, because of course that comes under the Health and Safety at Work etc. Act, which covers a much wider scope in this country than roads. It is worth asking the Minister what would happen if someone in control of these vehicles was actually found to be under the influence. Under what legislation would they be prosecuted, if they were liable?

The question of safety, as noble Lords have said, is fundamental. What worries me is that the Bill defines safety as meaning only

“to an acceptably safe standard”.

Acceptable to whom? What about the risk? Is there an acceptably low risk of committing a traffic infraction? Again, acceptable to whom? I am very concerned about the need, in all this legislation, to achieve a step change in road safety for all people who are affected by vehicles or what happens. At present, the risks of death or injury on our roads are significantly higher than for life in general or, indeed, on other transport networks, such as rail. Pedestrians and people who cycle —we have debated scooters before—bear a disproportionately higher risk of injury. If we add in children, old people and people with disabilities, who are particularly vulnerable, this is something that we do not really seem to take very seriously.

One issue that came up in a debate on the last group of amendments, which the noble Baroness, Lady Bowles, raised, quite rightly, was the question of testing on the road, but it is a question of “Which roads?”. Most people think that the first location for testing these vehicles will be on a motorway, because there are no pedestrians—or there should not be any pedestrians or cyclists there—and that is quite simple, really. But then, when we drill down, apart from motorways or dual carriageways, what other groups of roads would one have to test these vehicles on? It becomes very much more difficult and very subjective. I do not have an answer to this, but I am absolutely certain that the noble Baroness is right to say that it needs doing, and in a comprehensive way across all the different types of roads and tracks, in the countryside as well as in the towns. I am not quite sure where we are going to end up, because the amendments in this group on safety are fundamental. I do not have a detailed preference for which ones, but I am absolutely certain that we need to tighten up the definition of road safety to something that is not just acceptable but very acceptable, to a high standard, safely and legally.