Positions in Robotics at University of Sydney

In addition to the post-doc I advertised yesterday, two other positions might be of interest:

(1) Research Fellow in Robotics and Intelligent Systems

Centre for Robotics and Intelligent Systems
School of Aerospace, Mechanical and Mechatronic Engineering
Faculty of Engineering & Information Technologies
Reference No. 1632/0817

(2)  Lecturer / Senior Lecturer in Robotics and Intelligent Systems

Centre for Robotics and Intelligent Systems
School of Aerospace, Mechanical and Mechatronic Engineering
Faculty of Engineering & Information Technologies
Reference No. 1633/0817

Both positions include “Transport and Logistics” as a potential application area.

Postdoctoral Research Associate in Transport Networks. Closing Date 11:30pm 19 November 2017  

School of Civil Engineering
Faculty of Engineering and IT
Reference no. 779/0417A

  • Join an organisation that encourages progressive thinking
  • Be valued for your exceptional knowledge and experience in Transport Networks
  • Full-time fixed-term for 3 years, remuneration package: $106k  (which includes base salary, leave loading and up to 17% superannuation)

About the opportunity
Applications are invited for the appointment of one Postdoctoral Research Associate (Level A) in the School of Civil Engineering, within the Faculty of Engineering and IT at the University of Sydney. The position will support the research and leadership of School of Civil Engineering in the newly launched Transport Engineering program.

The successful applicant(s) will help build the new research group headed by Professor David Levinson to further the analysis of Transport Networks, understand the relationships between Transport Networks and Land Use, and consider the implications of changing Transport Technologies on optimal Network Structure.

Applicants should hold a PhD in civil engineering or a related field. They should be able to demonstrate high quality research in the area of transport networks, geo-spatial analysis, and econometrics. Demonstrated ability to publish research outcomes in high-quality international journals is also essential. Since the position will require frequent liaising with government and industry, applicants should demonstrate strong communication skills.

About you
The University values courage and creativity; openness and engagement; inclusion and diversity; and respect and integrity. As such, we see the importance in recruiting talent aligned to these values in the pursuit of research excellence. We are looking for a Postdoctoral Research Associate who:

  • has a PhD in civil engineering, or related fields,
  • conducts high quality research in the area of transport networks, geo-spatial analysis and econometrics,
  • demonstrates ability to publish research outcomes in high quality international journals, and
  • possesses strong communication skills as will be required to liaise with government and industry stakeholders.

About us
Since our inception 160 years ago, the University of Sydney has led to improve the world around us. We believe in education for all and that effective leadership makes lives better. These same values are reflected in our approach to diversity and inclusion, and underpin our long-term strategy for growth. We’re Australias first university and have an outstanding global reputation for academic and research excellence. Across 9 campuses, we employ over 7600 academic and non-academic staff who support over 60,000 students.

We are undergoing significant transformative change which brings opportunity for innovation, progressive thinking, breaking with convention, challenging the status quo, and improving the world around us.


The University of Sydney encourages part-time and flexible working arrangements, which will be considered for this role.

For more information about the position, or if you require reasonable adjustment or support filling out this application, Monika Browning, Lead Talent Acquisition Consultant, on +61 2 8627 6562 or monika.browning@sydney.edu.au

If you would like to learn more, please refer to the Candidate Information Pack for the position description and further details.

To be considered for this position it is essential that you address the online selection criteria. For guidance on how to apply visit: How to apply for an advertised position.

Closing date: 11:30pm 19 November 2017

The University of Sydney is committed to diversity and social inclusion. Applications from people of culturally and linguistically diverse backgrounds; equity target groups including women, people with disabilities, people who identify as LGBTIQ; and people of Aboriginal and Torres Strait Islander descent, are encouraged.

If we think your skills are needed in other areas of the University, we will be sure to contact you about other opportunities.

The University reserves the right not to proceed with any appointment.

Selection Criteria

Candidate Information Pack

To Apply, Go Here.

The quest to design a smarter road | fast co-design

Katherine Schwab in Fast co-design writes The quest to design a smarter road. I am quoted:


One key element of safety when it comes to connected anything, whether a smart fridge or a smart road, is how easy it is to hack. While 3M hopes to become the leading purveyor of connected roads, it’s still figuring out how to make its technology secure–complicated by the fact that national standards don’t yet exist for it.

According to David Levinson, a professor at the University of Sydney who leads the Network Design Lab and the Transport Engineering group, this is one of the biggest problems with any smart infrastructure. “Camera sensors and programs that analyze images can be spoofed by markings on stop signs to make them think they’re a different type of sign,” he says. “It’s going to be a cat and mouse game. If someone can broadcast the 55 mph sign, I can sneak up to the speed limit sign and broadcast that the speed limit is 95 mph.”

“We don’t have a standard for what would help the driverless car,” Levinson says. “We don’t have a standard for, this is the most efficient paint marking, or we’re going to put this RFID in the sign to broadcast this message.” That means it’ll be difficult for any state to begin implementing technology when there aren’t any guidelines for what to invest in–a significant barrier for 3M as the company looks to market its new products.

On Indulgences and Carbon Offsets

Catholics have a notion called “Indulgences”. Wikipedia summarizes it as “a way to reduce the amount of punishment one has to undergo for sins.” In the Middle Ages, indulgences were commercialized, so wealthy people could buy themselves out of punishment (or the loss of wealth might be considered the punishment, if you want to be charitable).


In the modern world, carbon emission is a great sin. Those traveling by air sin the most. Prominent environmentalists are often targeted for hypocrisy, and those hoping to avoid hypocrisy might purchase “Carbon Offsets“. These are the more equivalent of commercial indulgences. The airlines offer them to guilt-ridden passengers.

Perhaps the most obvious, ‘common sense’, solution when demand (pollution) is in excess of supply is to expand capacity.  This is what we do with most things if we can. If our house is too small, we make it bigger. If our wallet can’t hold all of our cash and ID cards, we get a bigger one. If the internet is too slow, we add capacity. In roads, this usually means adding lanes to existing roads. Perhaps we could plant more trees to absorb more carbon pollution.

Consider for instance the Boston to London round trip. It is 3255 miles each way (5237 km) and 1.1799 metric tons of Carbon roundtrip. For $14.16 or 1,888 Award Miles a United Airlines passenger can support the  Alto Mayo Conservation Initiative. Objectively this is not a lot of money in the scheme of things, and maybe it will offset your trip. I don’t have the impression most travelers purchase these indulgences.

More importantly, I don’t think this scales. Some estimates below:

A Trans-atlantic flight might require 11 trees per person per flight to do a full offset.  There are about 100 million international enplanements from the US per year.  Not all are Transatlantic of course, many are Trans-Pacific or to South America, and so longer. I will leave it to a research paper to figure out total distance. So that is on the order of 1100 million trees per year (probably more) to be planted to guiltlessly offset US international air travel.  Let’s assume 5m x 5m per tree (25 m^2). 25*1100M = 27,500 million square meters to offset international aviation from the US (excluding US domestic aviation and travel in other countries. That is 27,500 km^2, or an area of about 165 x 165 km on edge per year (for say 50 years until aviation switches to biofuels). This is the size of Massachusetts.

While that is technically feasible, since the US has lots of land (and is more than 50x the size of Massachusetts, as Massachusetts is a smaller than average state), no-one is actually doing this, and the offset is over the life of the tree, not immediate, so we would need one Massachusetts per year until the end of carbon-emitting aviation to make offsets work.

I like to think in terms of queues. The environment can clear (absorb) a certain amount of CO2 per year, basically the equivalent of net zero carbon emissions. If there is a positive amount of emissions, the CO2 queues in the atmosphere, waiting to be absorbed. (And probably doing things to the environment we wish were otherwise.)

If offsets are not employed, the alternative is that the accumulated CO2 queue from US Transatlantic enplanements will continue to grow. We could pull out Kant’s categorical imperative “Act only according to that maxim whereby you can, at the same time, will that it should become a universal law.” and argue since this doesn’t scale (can’t become a universal law), you shouldn’t do it. But that’s the sort of philosophical nonsense that we hope philosophers have recovered from.

Just because it can’t solve the entire problem and can’t become universal doesn’t mean it can’t be useful to plant more trees.  Trees are good. However, while a carbon offset indulgence may absolve you from guilt on a particular trip, it cannot absolve the industry, since it cannot scale.  Imagine the number of trees required for all aviation, not just international, and for auto travel (about 10x aviation). A more serious solution is required, one which either takes CO2 out of the air more efficiently, produces less CO2 per flight (through say biofuels or electric power), or reduces the number of CO2 emitting activities like flights (and internal combustion engine car trips) (by reducing travel).

Now to be clear, if you expand the capacity of the planet to absorb pollution (i.e. plant more trees), and people pay for their pollution, the reduced cost of per unit of pollution means that people will pollute more. Drivers will travel longer, industry will use less socially efficient means for energy generation. There might be a small amount of GDP growth associated with both the geo-engineering and resource extraction, so it is not entirely a bad thing, but it may not solve your pollution problem.


The Transportist: September 2017

Welcome to the September 2017 issue of The Transportist. As always you can follow along at the blog or on Twitter.


Transportist Posts

Media Appearances

Transport News





EVs / Energy

HPVs / Bikes

SVs / Taxis


Land Use

Equity / Justice



no hyperloop news this month

Retail / Delivery


Aviation and Space




By Us

Spontaneous Access: Reflexions on Designing Cities and Transport – Now in print

Spontaneous Access: Reflexions on Designing Cities and Transport is now available as a high-quality color trade paperbackIt is also available in hardcover.

It remains available as an eBook on Kindle Editions  and at the iBookstore. If you have the option, I encourage you to get it on Apple’s iBooks, where it has additional features, like pop-up references and image galleries, as it was designed in ePub3.

Spontaneous Cover-Front

Table of Contents

  1. The City Spontaneous
  2. The 60-Year Line
  3. Community without dendricity
  4. The pint-of-milk test
  5. The timeless way of building networks
  6. Axioms about roads
  7. Garden streets
  8. Vitality
  9. An archipelago of walkability
  10. Filling in
  11. Leapin’ frogs
  12. The reorganization of road function
  13. Beyond the plan view
  14. Interfaces of freedom
  15. Instruments of control
  16. Shared space
  17. Winter is coming
  18. Diversity as insurance
  19. Differentiate city and country
  20. Don’t confuse the place for the time
  21. Great Britain doesn’t have an Americans with Disabilities Act
  22. Designs serve varied and sometimes conflicting interests
  23. A vision of visions
  24. A faster horse
  25. The Ant and the Grasshopper
  26. Deconstructing Busytown
  27. Spontaneity in a can, spontaneity in a plan
  28. Building the city spontaneous
  29. Framing regional development
  30. First do no harm

There are several themes in the book:

Cities and their networks operate on multiple timescales simultaneously. Traffic lights change by the second, rights-of-way last millennia. Cities see massive daily flows of people in and out. The core, timeless, enduring elements contrast with the faddish ephemera that too much effort is focused on. The future is emerging, but determining what we are looking forward to will be enduring or ephemeral should be the critical focus of anyone involved with transport and city design.

This book does not shy away from the normative and prescriptive. In this it differs from much academic work, including my own, which tends to the positive and descriptive. Principles are laid out, which I believe to be true and correct, many of which are not scientific in the way they are framed. They of course may lead to testable hypotheses, but they are also value-based.

The idea of the ‘spontaneous city,’ one that serves needs and wants in real-time, is a theme running through both the title and the text. What conditions encourage people to take advantage of their city (and therefore make it stronger)? What conditions worsen life for the users of the city?

The emergence of new transport technologies gives us a chance to restore and correct, to right what is wrong with the places we live. From the railroad and electric streetcar creating separation between places where people lived and where they worked, to the elevator enabling high rise construction, to the motorcar which put suburbanization into over- drive, all significant transport innovations reshape cities. The new autonomous vehicle, the new electric vehicle, the new shared vehicle, the vehicle form, shape, and size are a transformation of similar scale and scope. These changes will create opportunities over the coming decades, which we can seize or reject.

This book is about how cities do work, how cities can work, and how cities should work. In part it is about traditional fields of planning and engineering, but takes a much broader concept of design principles than those fields usually do. This is because it is also about evolution and it is about opportunism. The world is changing fast. We can make it a more humane place than it has ever been, or we can allow it to devolve into a more brutish environment, where we remain a victim of our collectively built environments, rather than their master.

When the book speaks of ‘cities’ it really means the entire metropolitan ‘urban system,’ not just the historic core city (or the central business district). Downtown is but a part of the city, and the central city in many metropolises is not even a plurality of residents.

Much of this book includes complaint, and it may feel like shouting into the wind. But every complaint is about a design failure, either with intention or by accident, that degrades experience for everyone, or degrades the experience of some for the benefit of others. Life is comprised of tradeoffs, but not all tradeoffs are made at the appropriate rate of exchange. Both cities and their transport networks are the product of thoughtful human actions and unconscious emergent processes, where systematic behavior drives the underlying logic of designs.

The optimal design of transport networks to serve the goal of spontaneous access cannot be determined in the absence of knowledge about the actual development pattern. The optimal development pattern cannot be known without regards to the plan of the network. Discovering the right combinations of networks, land use, and other urban features is what makes cities successful. The measure of their success is their population, their wealth, their happiness.

But even more importantly, the optimal transport network for the technology of one era is not necessarily the optimal network of the future, and the same is true for development.

Much of Spontaneous Access is drawn from my blog transportist.org, or streets.mn, although it has been significantly edited and reorganized from posts that may have appeared there. In that sense, it is a younger sibling to the recent (2015) book The End of Traffic and the Future of Transport with Kevin Krizek. It is a collection of reflexions (a somewhat archaic British way of spelling “reflections”), short essays that collectively give insight into today’s design problems and some possible solutions.

How you want your books

Since I am preparing several manuscripts for release soon, I did a series of Twitter polls on the preferred format of release. The results are as follows below. While these are not ‘scientific’, unlike say, nothing, they seem right, and while results no doubt vary by field, this seems mostly right.






In short, for paper, you clearly want softcover AND color. You probably want cheaper paper, but I suspect this really depends. A slim majority prefer PDF to ePub format at the same price.

So you will get softcover AND color, and I will fret about whether the paper/image quality from the higher cost is worthwhile. If there is demand, there may be hardcover too, but I don’t think anyone but my mom and maybe a library will get it.

You will get PDF, and maybe ePub too if I’m feeling either generous or obsessive.


A Busful of Knowledge: On Going from the University of Sydney to the University of New South Wales

There are a few universities in Sydney, the University of Sydney of course, is the most important (revealing my bias), and the University of Technology Sydney (UTS) is nearby, but the University of New South Wales (UNSW) also has a very strong reputation in Civil Engineering and overall, and really isn’t that far away, nothing compared to Berkeley vs. Stanford. Together they can and should make this region a technology powerhouse in the same vein, if not to the same extent, as Silicon Valley or Boston’s Route 128.

The universities are only an hour apart by walk (17 minutes by car on the shortest walking route), but the transit connections are circuitous and nearly 40 minutes. Look at the images. On the top is the relatively direct walking path, just a bit over 5 km. On the bottom is the shortest travel time path by transit, 37 minutes and far out of the way.  Perhaps if we want to encourage interaction, we should make interaction easier. Keep in mind not everyone has a car, especially students, and especially people who commute by train or bus, and taxis are still expensive.

Imagine how much more powerful the region could be if the researchers could more seamlessly interact and students easily cross-enroll and attend classes. Now I realize these are rival institutions established to be competitive, so there are limits to getting along. On the other hand, a more direct bus route seems like a good place to start.

Infrastructure Spending Done Right

I wrote a piece for the Van Alen Institute: Infrastructure Spending Done Right. This is part of a wider Van Alen Report: America’s Infrastructure,  including articles by

Infrastructure Spending Done Right

Van Alen Institute: America's Infrastructure. Report 19.
Van Alen Institute: America’s Infrastructure. Report 19.

Infrastructure spending as stimulus appeals to politicians and voters because it would appear to kill three birds with one stone. Ostensibly, critical infrastructure is repaired or newly constructed, job opportunities are created for the unemployed, and the greater economy is set on course for growth. But how and where funding is spent frustrates these objectives. Federal funding often winds up disproportionately in rural areas at the expense of dense, growing cities where long-term economic benefits would be greater. Moreover, job creation is dubious given the high level of skill required for construction work and increased role of technology on the project site. Although greater investment in maintenance could both give relief to the unemployed and boost the benefits of existing infrastructure, politicians eager for ribbon cutting ceremonies often choose new infrastructure over repairing the roads, bridges and railways we currently have. David Levinson, transportation expert and professor at the University of Sydney, takes us though past and potential future infrastructure spending initiatives, and explains how setting the right priorities can ensure our infrastructure provides greater prosperity over the long term.

On December 6, 2008, in the throes of the Great Recession, then President-elect Barack Obama laid out key parts of his Economic Recovery Plan. In his radio address he boldly said “ … [W]e will create millions of jobs by making the single largest new investment in our national infrastructure since the creation of the federal highway system in the 1950s… If a state doesn’t act quickly to invest in roads and bridges in their communities, they’ll lose the money.” This plan turned into the American Recovery and Reinvestment Act, with a total budget of $831 billion. It dedicates $105 billion to infrastructure, of which $48 billion went to transport.

The value of the projects from the 2009 stimulus remains questionable. Projects tended to fall in rural areas (mostly road resurfacings) not because the work was essential, but because they were “shovel-ready” and easy to do. The projects were easy because they were already designed and had environmental permits in place. But the fact that these projects were so far along in development, yet remained unbuilt, suggests that they were not the highest priorities for the local and state transportation agencies that oversaw their construction.

Though administrations have changed, the disproportionate allocation of federal spending to rural areas over more developed cities – where the majority of needed infrastructure work exists – will likely go unabated. President Trump has proposed various tax incentives to stimulate $1 trillion worth of private investment towards the nation’s infrastructure. While Trump has discussed urban-based projects, such as rebuilding New York City’s poorly-managed airports, the Republican party – which he leads but counts as its base mainly rural voters – will likely exacerbate the overfunding of rural projects even more, if only to get its own representatives and senators re-elected.

Along with project location, setting job creation as an objective of infrastructure spending can also undermine the economic value of projects. At the time of the 2009 stimulus, unemployment was around 10%. With more workers looking for jobs, spending on infrastructure during a recession may arguably bring labor off the sidelines, while also taking advantage of the temporary wage drop due to the joblessness spike. In short, the state can get more infrastructure built for less, and put people to work who would have been otherwise unemployed. Today, however, unemployment is around 4.7%. Competition for labor is up, and with it construction wages. And without slack in the labor market, new projects are more likely to shift employed workers around, not add new jobs to the economy. Worth noting is President Trump’s assertion that his proposed tax breaks will pay for themselves. If these privately-funded projects fail to increase the net number of jobs, the hope for additional revenue to offset tax incentives will never come into reality.

Further complicating the job scenario is the capital-intensive nature of construction today. Macro-economists or policymakers who think of highways and transit lines as engines of job creation are remembering grainy black and white images of Civilian Conservation Corps workers slinging pickaxes as they build roads through national parks. Construction projects are more capital intensive than they were in the 1930s, using heavier machinery and far less labor. As technology advances, and construction equipment becomes increasingly roboticized and automated, jobs will become highly skilled and decrease in number. Most infrastructure construction jobs already require two or three years of apprenticeship and on-the-job training. In the future, infrastructure stimulus may offer little for unemployed people without extensive construction experience.

While the creation of jobs from infrastructure construction is limited, there are potential long-term benefits of constructed infrastructure in terms of jobs. It is worth noting that our current surface transportation system is not just in need of repair. In most parts of the U.S., our system connects everything worth connecting, and does so as cost-effectively as possible. There’s little need for new infrastructure, but great urgency to rehabilitate the infrastructure we already have.

Local and state governments are largely responsible for preserving existing infrastructure. They can use additional federal support. But we should be sure that any support is pushed toward maintenance, not new infrastructure which largely serves as a distraction. We all know that maintenance, repair, and reconstruction are not sexy. They do not result in ribbon cuttings with smiling politicians getting their pictures taken and posted in the local news. Yet on a per-dollar basis, fine-grained maintenance work employs more people than large-scale greenfield construction. Moreover, it is ideal to run the capital equipment required for road construction at a continuous level, thus maximizing its productivity. Continuous utilization is achieved by a steady rate of spending on projects, not stimulus-related spikes or failures to authorize infrastructure expenditures.

Economic activity increases with accessibility – more specifically, the ability for workers to reach jobs and stores, and for firms to easily interact. This occurs with faster and more direct transport, denser land use, and increased access to developed urban areas over less economically active rural areas. That said, it is cheaper to build in rural areas than cities, so the cost-to-benefit ratio is not obvious. This ambiguity is worth noting. While infrastructure policies may aim to even out spatial inequities and “spread the wealth,” that ambition is at direct odds with the desire to maximize the productivity and efficiency of infrastructure.

Public works are justifiable when social benefits exceed costs, not because they create spikes in job growth or score political points. To maximize the amount of infrastructure society gets per dollar, the government needs to be efficient about how infrastructure money is spent. From an infrastructure perspective, if a road project employs some people, that provides a nice rhetorical flourish; but if projects are aimed solely to employ people, the state will be wasting money which in the long run shrinks the economy. The debt borrowed to build said projects ultimately comes due.