The Rise of the Robo-Bus

Published on Linkedin 26th April 2016.

The world is getting excited about autonomous cars, in addition to the application of autonomous driving technology to cars, there are many opportunities in heavy-duty vehicles. A major example of this is the humble bus, the backbone of our public transport infrastructure.  The adoption of autonomous driving technology to public transit vehicles will drastically change the public transportation industry beyond recognition and it seems to be gathering some real pace. This week some news reports that Tesla was working on something that was going to shake up Public Transport. That has left a few bus manufacturers scratching their heads and worried.

Operating buses is a huge industry worth $Billions on a global basis annually and often receiving heavy government subsidies as an essential service. In the worlds growing cities buses play an essential role in public transport and good bus services are critical to reducing congestion and improving air quality, even with diesel engines a well designed urban transit system using buses is still better than lots of individual car journeys from an emissions and congestion point of view. There will be major differences between an autonomous bus in the future (and my prediction of what Tesla have in-store) and what we are used to seeing today:

1. It can be smaller than current transit buses – In a bus operation, the major costs to the operator are the capital cost of the vehicles, the cost of their staff and then running costs like fuel. Sized for peak capacity requirements to deliver the most payload for the smallest operating cost, it is not uncommon at off-peak times to see buses running under capacity or completely empty. Removing the need for a driver will allow a much greater level of flexibility to be designed into the delivery of these services. Whilst there will still be a need for some larger vehicles to maximise capacity at times of peak demand, the overall trend will be for a greater number of smaller vehicles. For example in the last couple of weeks, we saw the launch of the WePods in the Netherlands, running fully autonomously on Elektrobit Software and using Mapscape mapping. This cute little shuttle bus is also using an all-electric drivetrain. Note the WePod is quite small, or the similar-looking EZ10 from Easymile.

2. It will be connected – the autonomous systems will be connected to real-time passenger information systems feeding live information about the location of vehicles to passengers, nothing new there, but the systems will also be more closely linked to traffic control infrastructure and other vehicles. Meaning smoother traffic flows in the city and fewer snarl-ups. The benefits of connecting the vehicles to each other can be seen in the recent PR around the EU funded truck platooning project showing the ability for the trucks to safely “tailgate” each other at motorway speeds and come safely to a stop. Apply this to buses in city driving traffic and you have a demand-responsive bus service, which can save road space and be much safer.

3. It will be safer – the latest advances in driver aids for large vehicles in city traffic are using the same sensor systems as autonomous driving systems to provide information to the driver about the location of pedestrians and cyclists. The issue with many of these systems is that additional information can be distracting to the driver with sensory overload. Fully autonomous vehicles can “have eyes in the back of their heads” with 360-degree sensors and see around corners. Increased hazard perception and faster reaction to hazards, with no lapses in concentration, will result in far fewer accidents.

4. It will be cheaper to operate – there are some obvious ones here, buses like other large vehicles are difficult to drive, you need special training to be able to operate them safely and also ongoing training and testing to ensure that the driver remains competent. No drivers, no special testing, no ongoing training, and testing leading to greatly reduced manpower costs. The ability to have a more demand responsive system where the capacity is flexed during the day means less transporting boxes of air. Adopting autonomous systems will also mean less wear and tear on the vehicle components as the driving will be more consistent and improved demand responsiveness will result in greater system spare capacity. Fewer collisions mean less damage to repair. A major cost in delivering a bus service is what is sometimes called deadheading, which is where the bus has to go from its depot where it is parked up overnight to the area where it actually is running its route. The cost of making this trip in terms of fuel and manpower means they are typically minimised in current bus operations. An autonomous vehicle will completely transform the economics of vehicle operation in more ways than one.

5. It will be electric. Smaller vehicles mean more manageable electric vehicle powertrain and practical battery packs. Electric vehicles are desirable in urban situations due to no emissions at the point of use and low operating noise. These vehicles will be clean and green! EV also suits start-stop operation and application-specific drivetrain like a low-speed bus is relatively straightforward.

So some clear benefits there, what about the downsides? Like all automation, there is a human cost in terms of lost employment opportunity, there will be new opportunities in the design, development and ongoing maintenance of the vehicles, but this will not fully offset the reduction in employment that is possible.

So why Tesla?  Whilst electric drive is not essential for autonomous driving, the two things do support each other very well. It is likely that autonomous driving technology will be a facilitator in the increase of electric buses. It will all be about the control algorithm and the data. Google and Tesla are for instance collecting vast amounts of data now to build control algorithms using a technique known as machine learning. These developed control algorithms will be valuable. As mentioned in a recent post from me about the BYD electric buses in London, an electric bus needs a lot of batteries to be able to deliver a practical range/payload and that is good if you happen to own a battery manufacturing plant.

Surprisingly to me a lot of people seem to think this is all going to take more than their lifetime to become a reality, the technology is getting closer every day. Check out the truck platooning videos, the WePod, and Easymile it is here now and there are also others out there. It is close to being a practical reality and will gain momentum rapidly, especially as the costs of the technology to enable it all is falling.

Meanwhile, in China, search engine Baidu is also developing self-driving buses, they are saying only 3 – 5 years till a commercially viable product is ready…

About the author; Ryan Maughan is the Managing Director of the AVID Technology Group Ltd. AVID is based in the North East of England and is a leader in the design and manufacture of technology for electric, hybrid and conventional heavy-duty and high-performance vehicles.