How Will AVs Navigate Signalized Intersections?

Autonomous vehicles promise such a degree of improved safety that we sometimes picture them as all-powerful, faultless machines. To a large extent, they will need to be. But how can we be sure they’re safe even in the most risky situations?

Consider the signalized intersection. Since their appearance at the end of the 19th century, traffic lights have been the primary mode of granting access to road intersections. However, traffic statistics show that, despite their claim to only a tiny percentage of road area, intersections are where 25% to 45% of all traffic collisions occur. Why is this?

When you think about it, it starts to become obvious. Intersections bring together cars traveling in all directions and then asks them to proceed through in groups timed by traffic signals. If any one car makes a false start, or speeds through the intersection and collides with a car in front of them, the entire exchange is compromised. And since cars are coming from all directions at various speeds, the risk of an accident that causes injury and/or major damage to vehicles goes way up, too.

So how will autonomous vehicles navigate the signalized intersection?

Therein lies the debate: will AVs independently “sense” their way through intersections using onboard sensors capable of discerning, for instance, the color of traffic lights, all while communicating with other vehicles (V2V) — or will AVs be “escorted” through intersections via connected infrastructure (V2I)?

As it turns out, the case for the V2V argument is in question. For starters, we would need to trust that every manufacturer has built in precise, latency-proof communications units that ensure constant communication with every other car. Statistically, the odds of that happening aren’t good, and it seems like too big of a risk when all it takes is the failure of one sensor on one car to cause an accident.

Furthermore, according to a paper by the Journal of Artificial Intelligence Research, “a vehicle approaching an intersection can quickly find itself in a situation where a collision is unavoidable, even when it has acted optimally.” In other words, even if no car overtly malfunctions, you can hardly call the intersection safe. As the JAIR paper notes, an intersection has very specific points of potential collision – hotspots that only one car at a time can occupy. That’s engineer-speak for “two cars occupying one hotspot will cause an accident.”

And this is why there’s a strong case for the V2I argument (where AVs get escorted through the intersection). The paper goes on to argue that an intersection needs its own “cooperative vehicle intersection control system (CVICS)” – hardware and software located at every interchange, which keeps an eye on all the points of potential collision and guides AVs through based on its safety-first algorithm.

Remember that acronym, CVICS, because it’s going to be a huge factor in the success of autonomous vehicles. It is absolutely essential for the safety of AVs, which can’t reasonably be expected to never run into trouble in a complex intersection, no matter how advanced they become.

Researchers at JAIR aren’t the only researchers arguing that CVIC systems are going to be a necessity for autonomous vehicles. There seems to be a consensus forming around the idea, to the point where researchers are now trying to determine the best algorithms for intersection control. MIT recently reported research on a “slot-based system”, similar to airplane-boarding procedures, designed for slow guidance through intersections in groups. They were essentially building on an IEEE research paper that offered a detailed algorithm for intersection control units to deploy.

The results from IEEE’s testing of their algorithm is worth a look. Not only was the flow through the intersection optimized for safety as expected, researchers also noticed some other improvements: vehicle stop delay time was reduced by an astonishing 99%, improving average travel time (under the presumption that a commute will involve several encounters with CVICS) by 33% or more. Carbon dioxide emissions were reduced by 44%, correlating directly with fuel savings of 44%. Basically, the researchers found that CVIC systems not only ensured safety, but made the process of getting through an intersection much faster. What’s not to like?

We remain optimistic about the safety benefits offered by autonomous vehicles. But the research is out there, and it’s compelling. Rather than risk public distrust in autonomous vehicles following a few crash-causing anomalies in intersections, it makes sense to deploy CVIC systems with proven algorithms to make intersections foolproof.

Just as we will eventually cede the wheel to our AVs, we will also need them to cede the wheel to CVICS from time to time and, in principle, the statistics showing such a disproportionate number of accidents occurring in intersections will shrink down to zero. We know that in the future our cars will be smart, but their interoperability with safety-oriented systems like CVICS will allow them to be brilliant.

Rob Fischer is President of GTiMA and a senior advisor to Mandli Communications’ strategy team. GTiMA and Mandli Communications are both proud partners of the Wisconsin Autonomous Vehicle Proving Ground.

Follow Rob on Twitter (@Robfischeris) and Linkedin.

The Time Is Right For ADAS Testing

Have you seen this commercial? It starts with a woman in standing in the middle of the street, gazing up at a sky-written marriage proposal. We see that she’s in an urban area. A car comes quickly into the frame and even more quickly stops, as the words “automatic emergency braking” appear on-screen. Then the camera pulls back to show that we’re in a car going quite fast on a desert highway, with a couple watching this demonstration of emergency braking on their smartphone. “Wow, good to know we have automatic brakes,” says the woman. The man, entranced by the video, swerves out of his lane, which is corrected. “We have lane departure assistance, too,” he says with a smile.

There are a lot of problems with this ad, and those of us that are excited about the AV age should be paying attention to it and ads like it. Because the fact is the public does not currently trust these automatic driver assistance systems (ADAS). J.D. Power recently released the results of a new survey that shows “basic semiautonomous features such as adaptive cruise control, autonomous braking and lane-departure assist notched the largest increase in complaints this year.” The analysis argues “these complaints signal an uphill battle … if car buyers aren’t convinced that the basic building blocks of self-driving cars work seamlessly now, they won’t buy a self-driving car in the future.”

That should concern anyone with a stake in the ultimate success of autonomous vehicles. And the public may be correct to not trust them at present — these AAA statistics show that the efficacy of automatic braking systems is iffy at best. Automatic emergency braking, for example, only has a shot at working if the driver is going fairly slowly. There’s little consensus over what sensors should be used for which systems — that conversation is still ongoing — and there are other unknowns when it comes to ADAS that suggest many of these systems are not ready for prime-time.

So why are car companies pushing these systems so quickly? The ad described above shows they’re aware of the limitations of their systems. The car that narrowly misses a pedestrian is going slowly, but we immediately cut to a car that’s going quite fast — a sneaky way to make the AEB system appear useful on the highway, when according to AAA, it’d be somewhere between 9 and 40% effective when traveling at 45 miles per hour.

These companies are caught in an awkward position. They are aware of the need to maintain public trust once it’s time for level 3 and 4 autonomous vehicles. But they also need to continue to turn a profit in the meantime. So we can understand why they’re anxious to get these systems to market, but the systems need more than the small-sample public testing AAA carried out. There is still debate, for example, on whether lidar or radar is more effective for automatic emergency braking. Doesn’t it seem like this should be discerned before cars touting the feature are deployed? Shouldn’t consumers get systems that work as advertised? To ADAS our way to AVs, shouldn’t consumer trust be at least as valuable as a car sale?

We understand the current plight of the OEMs — the genie is out of the bottle, ADAS systems are on the road, and they are features that have been proven to be desirable, at least in theory, to consumers. Nobody wants to get left behind when it comes to new and cool features. This is where the Wisconsin Proving Grounds comes in. By testing ADAS systems at the Proving Grounds, OEMs can simultaneously ensure the safety of their systems and — since they’d be doing so out in the open — gain public trust that is so crucial to their business model, both currently and in the near future.

The time is right. It’s about to get easier to “ADAS our way to AVs” in terms of testing AV systems without needing an entire autonomous prototype. According to recent reporting, the House and the Senate are both advancing legislation to make AV testing easier on manufacturers, following hearings and meetings with over 100 industry stakeholders. There are over a dozen bills advancing through committees right now, but as far as the proving grounds go, we’re thinking this will help speed things along: lawmakers are looking to “raise the federal cap on the number of exemptions that can be granted to driverless carmakers who want to design and test cars without traditional automobile features.” Currently, while all cars are required to have steering wheels and pedals, federal waivers can only be granted a certain number of times per year for testing purposes, which constrains the ability of manufacturers to conduct serious testing. With that regulation eased, we think we’ll see a lot of ADAS action at the proving grounds.

So, our message to auto manufacturers: this is your moment. Public opinion may be turning against some of the early iterations of autonomous features in vehicles, but there’s a clear path toward righting the ship. It’s coming at just the right time, with a confluence of regulations and test site availability opening the doors to a new intensity in regimes for testing ADAS systems. And with a growing public attention to all things AV — which is sure to result from the big legislative push that’s ongoing — it seems pretty clear the public will take notice. When that starts to happen and testing shows, for example, improvements to automatic emergency braking systems, those commercials are going to connect as well as they were meant to in the first place.

Heath Davis-Gardner is a professional writer and editor who currently serves as Strategic Communications Specialist at Mandli Communications.

Why the Breakthrough of Blockchains May Be “Blockchainless”

Blockchains, or more generally, Distributed Ledger Technologies have a huge role to play in the future of mobility. With the advent of connected autonomous vehicles as well as the rise of electric mobility and shared usage models, Distributed Ledger Technologies and the future of mobility is a perfect match.

It is no surprise that the New Mobility ecosystem – automotive players, mobility service providers, insurance companies, telecommunications providers and smart city leaders – are exploring blockchain technologies. Many leading players are starting to embrace the key concepts of blockchains in order to solve real world problems related to identity management for people and machines, data control and integrity, information sharing, as well as trusted connectivity and cyber security.

However, it appears that blockchain technologies have not been able to move beyond proof of concepts. Production deployments have been limited to ‘private” or ‘consortium-based’, rather than public blockchains. There are many reasons that help to explain this statement. For one, the concept of blockchain technologies is still relatively new. Along with this, the notion of distributed peer-to-peer architectures and new business models is challenging conventional thinking. The main reason, however, is much simpler: Blockchain technologies have inherent limitations.

These limitations include lack of real-time validation of transactions, and the scalability issues that are inherent in the architecture. Related to this issue are transaction fees and the concentration of mining power.

The notion of blocks being generated through the consensus process makes real-time validation and scalability hard to achieve. In fact, it is inherent in the architecture. Transactions and the consensus process are sequential. Tens, if not hundreds of thousands of unconfirmed transactions in the case of Bitcoin introduce serious limitations for use cases. Furthermore, the concept of mining and the need for stakers to achieve consensus and create blocks introduces transaction costs. As complexity grows an increasing amount of energy and computational power is needed. Bitcoin’s mean transaction fees have already risen above $1. Even worse, transaction costs are difficult to predict and therefore undermine business models in the Internet of Things. In a world of micro-transactions the question arises, “who is going to pay for it?”

Why a Blockchainless Approach Might be the Future

With the IOTA protocol, an interesting new approach to distributed ledger technologies is emerging that does not rely on traditional blockchain approaches. The 4 co-founders behind IOTA – David Sonstebo, Dominik Schiener, Sergey Invancheglo and Serguei Popov – have been part of the “Blockchain 2.0” revolution and made major contributions to the space. For instance, Sergey Invancheglo invented proof of stake. As part of many projects they have experienced the inherent limitations of traditional blockchains and have conceived a new approach to distributed ledgers. The idea of IOTA was born in 2015.

The main innovation behind IOTA is that it is a “Blockchain without the Blocks and the Chain”. The Tangle is a new distributed ledger architecture based on a Directed Acyclic Graph (DAG) that shares the same underlying principles and the key benefits of blockchains, just without their inherent limitations. Let me explain how.

Rather than rely on the creation of blocks that are chained together, in the Tangle, a single transaction references two past transactions.

The process involves 3 simple steps:

1. Signing: The transaction inputs are signed with your private keys.

2. Tip Selection: Markov Chain Monte Carlo (MCMC) is used to randomly select two tips that represent unconfirmed transactions, which will be referenced by your transaction.

3. Proof of Work: In order to have your transaction accepted by the network, you have to do some proof of work.

Once these 3 steps are completed, the transaction will be broadcasted to the network. Another network participant will come along, choose the transaction in the tip selection process and validate it.

IOTA validates transactions without the need for miners and the associated cost of creating consensus. Instead of relying on a small subset of the network (miners / stakers), IOTA relies on the entire network of active participants to approve and validate transactions. Consensus is no longer decoupled from transactions, but rather an intrinsic part of the process. This allows IOTA to scale without any transactions fees.

The benefits of this approach are apparent. IOTA achieves a step change in performance and scalability, which makes it particularly suitable for the Internet of Things with ultimately billions of connected things. Given that there is no such thing as ‘always on’ connectivity in the real world, the Tangle also supports offline transactions through partitioning. This allows clusters of IoT devices, such as cars and trucks, to branch off and still make transactions using P2P communication protocols.

Since the transaction posting and consensus is parallelized, the network is inherently growing and scaling with the number of transactions. With the more transactions are being made, the Tangle becomes even more secure and efficient.

The creation of IOTA formed the backbone of the Internet of Things and the Machine-to-Machine (M2M) Economy: a world where scalability is a must, and where transaction fees are prohibitive to making business models work.

The fact that IOTA does not introduce transaction fees makes it uniquely suited to IoT and M2M where a large number of micro- or even nano-transactions can be expected. The IOTA team envisions a future in which machines trade resources such as computation, electricity, storage, bandwidth, data and much more without the involvement of any third party. IOTA opens the doors to an Internet of Things that is not only secure and scalable, but makes business models in the M2M economy practical. It sets the industry on a path where it can move beyond proof of concepts and instead to real deployments.

If you look up “IOTA” in the dictionary, you will find it is “something very small.”. If you “don’t care one IOTA about something”, it means you don’t even care about it one little bit.

IOTA is a blockchainless approach to distributed ledgers. Should you care about it? Given its huge potential to become the backbone of the Internet of Things, I predict that IOTA will turn into something very big. We should not only care about IOTA. We should all get involved and use this revolutionary technology to shape our collective future.

Click here to view a short introductory video, “Welcome to IOTA Universe”.

The Future We Want – Why I Am Passionate About Blockchains

Back in December 2000 I had my first meeting in my new role to help SAP develop a future vision and strategy for supply chain management. Together with Albrecht Diener I met Steve David, the then CIO of Procter & Gamble, who introduced us to his vision of a Consumer Driven Supply Network powered by a next generation Auto-ID technology and agent-based optimization.

My next trip was to meet the Director of the MIT Auto-ID Center, Kevin Ashton in Cambridge, MA. Kevin is credited with inventing the term “Internet of Things” back in 1999 and has played a key role in shaping it. Next to the MIT Auto-ID Center, Steve David also introduced us to Stuart Kauffman from the Santa Fe Institute, the godfather of complexity science. Together with brilliant minds like Bill MacReady, Tony Plate, Fred Seibel, Brian Birk and Brian Potter we looked at swarm intelligence and how ants and termites organize themselves, how birds flock and how schools of fish protect themselves from predators. Ever since I was fascinated by the notion of smart, connected things and a future of autonomous, self-organizing systems.

The challenge we faced was that we did not really have the underlying infrastructure to manage such systems. How do you manage a global network of smart, connected objects? How do you give them an identity? How do you steer the behavior of these independent economic agents in a way that leads to a somewhat optimized outcome for the entire system? How do you enable transactions between these smart objects? How do we create a record of transactions in a world of distributed systems? And how do you keep such a system secure? We could not really find a satisfactory answer to these problems at the time.

When I discovered blockchain technologies almost 15 years later, I felt like I finally found this missing piece. Blockchains or more generally distributed ledger technologies promise to deliver the backbone for the Internet of Things. In the not too distant future, we might have fleets of electric robot taxis swarm out like ants to look for passengers. Distributed ledger technologies will play a critical role in this emerging Machine-to-Machine economy.

We have done projects with world leading companies to explore the potential of the technology. The more we learn, the more excited we are about its potential to change the world for the better. We believe that distributed ledger technologies are critical to shaping our collective futures. Thanks to people like Rob Wolcott, Yossi Vardi, Maurizio Rossi and Plamen Russev we have become part of a global ecosystem of innovation. In our many discussions at events such as DLD, KINglobal, webit, 4YFN, Kinnernet, Autonomy and TU Detroit more and more people seem to agree that we need to make sure that technology serves us as humans. We all need to make sure that we create a future where humanity, life and the planet can thrive.

We aim to create a platform to empower more people to understand the technology and its key concepts and implications. In the next weeks and months, we will share more and more of our thinking around blockchains and distributed ledger technologies.

· We will make the technology and its key underlying concepts and implications understandable to a larger audience.

· We will share our passion for a world where people and things have digital identities that can be shared and trusted across a network.

· A world where our data is no longer compromised by storing it in large honey pots of data that attracts hackers.

· We will explore why distributed ledger technologies are critical to securing the Internet of Things. We will explore the role of blockchains in machine learning and artificial intelligence, autonomous robotic systems and the emerging Machine-to-Machine Economy.

· We will examine new business models and how we as humans can thrive in this future.

We want to build a future we all want to live in. I hope you will join us on that journey. We all must play an active role in shaping the future, or somebody else will create it for us. A future we might not like. Let’s create the future we want!

Alexander Renz is Director at the New Mobility Lab, a not-for-profit organization with the charter to raise awareness and scale knowledge related to distributed ledger technologies in the New Mobility ecosystem. New Mobility Lab is a platform for sharing real world experiences and executing pilots and proof of concepts across the public and private sector. To learn more, click here.

Busting The “Heavier Traffic” Myth

When you’re tracking the evolution of autonomous vehicles via Google News alerts, every day starts with something interesting and often exciting. But some of the time, an article is only interesting because of how misleading it is.

There’s a reason for that. As we’ve been reporting for some time now, sometimes a writer or a publication decides to take a sliver of information out of context and turn it into click-bait.

The result, of course, is a scary and/or discouraging headline blowing up your RSS feed. This week’s culprit: “Self-Driving Cars Could Be Terrible for Traffic – Here’s Why.”

This is a rare case. Not only is the argument spurious, the tone actually gets worse after the headline.

Usually, a click-bait headline gives way to an article that begins to pull its punches or makes caveats. But the lead three paragraphs of this particular article take this fascinating progression: “(p1)Self-driving cars might make your future commute a lot more pleasant, but they won’t eliminate traffic” … (p2) Execs like Sergey Brin of Google say they will… (p3) “But experts say the vehicles’ impact on traffic will be minimal or negative.”

So, it’s not really that they’re saying self-driving cars “could be” terrible for traffic. They’re saying they will be.

Of course, as expected, the expert quoted in the following graf says nothing that calls for that sort of language. The expert says “autonomous vehicles won’t fix congestion woes unless a pricing system is put into place” for zero-occupancy vehicles to discourage companies from using AVs as a free alternative to shipping carriers like UPS and the postal service.

Buried deeper in the piece, almost as an afterthought, are examples of states that have already begun putting such pricing systems into place.

It may seem like a small thing. This is just one article, after all, in a sea of articles, if you search “autonomous vehicles and congestion” that all say in one way or another how much AVs will help reduce congestion. ( Here’s one of my recent favorites, from MIT’s Technology Lab, about how just a few AVs on a road among standard-driver cars dramatically reduces congestion.)

But this is how myths are made. The power of the Internet is strong, and the power of the human mind to think it knows something because it read it on the Internet may be even stronger. For the average news consumer who hasn’t been making an effort to stay abreast of every new study and survey that point to the undeniable promise of autonomous vehicles, all it takes is chancing upon this one article, and suddenly they know a new “fact” about AVs.

And they tell their friends. And their friends tell their friends. And so on, and so on.

But as we argued months ago, this is a deeply irresponsible act every time a respectable news outlet plays on the public fears of the unknown to make a spurious argument just to get some traffic for their banner advertisers. To paraphrase Elon Musk in the above-linked piece, every time a journalist writes an anti-AV piece, they’re potentially putting human lives at risk.

Because we shouldn’t lose the big picture here. Yes, states will have to adjust regulations to account for businesses that want to use AVs as free courier services. The switch to autonomous vehicles isn’t going to happen overnight, and it’s not going to be a situation where we just have a new kind of vehicle but we keep all the laws exactly the same. But the idea that states, which often find it extremely hard to tighten the belt enough to find money for road repairs, would simply go along with companies using AVs for free deliveries is absurd. Of course the company would need to pay for using public infrastructure for wide-scale delivery services. It’s hard to think of a state that wouldn’t jump at the chance of opening a revenue stream for transportation in this way.

Here’s the big point this argument gets us to, though – one totally missed by its author. So let’s imagine this future where businesses are having to pay a per-mile tax for unmanned delivery services. First of all, they’re economizing their own use of VMT (vehicle miles traveled) to get the job done for as little as they can have to use the roads, which means the nightmare scenario put forth by this article never comes true (as the expert cited in the article says) and yes, traffic congestion does decrease. The state, meanwhile, is getting some extra money to make road infrastructure repairs and even invest in new pieces of infrastructure. The environment is benefitting from the drop in congestion, and maybe will even more from smart infrastructure the state invests in. And people are still getting their packages on time.

This is the thing with these misleading articles. If you stop to logically step through the scenario they’re positing, and you account for basic common sense they forgot to include, their hollow scare piece turns into another rosy-looking scenario for AVs.

This isn’t just optimism on my part. Here’s a little experiment you can do to see my point in action. Go ahead, search around the internet for the benefits of AVs on any given topic, from safety to land use. Notice the hundreds of immediately available news articles and the plethora of white papers and academic journal articles.

Then Google the exact opposite of what you just saw hundreds of articles about. Example query: “autonomous vehicles will be less safe”. Something we know beyond a shadow of a doubt isn’t true. And yet, there you see all the click-baity counterexamples.

If I’ve written in a colloquial style here, it’s only been because I want to keep your attention. The point is deadly serious. Journalists need to stop worrying so much about getting clicks and start being responsible to their sources as well as their readership by avoiding this sort of crass misappropriation of context.

News outlets, just consider it: it’s completely possible to support your business model without scaring people into clicking headlines. AVs are exciting enough in their own right. We know you’ll catch on soon.

Heath Davis-Gardner is a professional writer and editor who currently serves as Strategic Communications Specialist at Mandli Communications.