Re-Inventing The Wheel: The Impact Of Autonomous Vehicles On Civil Engineering

Autonomous vehicles are being developed for our cities and roads and will create a fundamental shift in vehicle use and infrastructure design. These vehicles are proposed as safer and more efficient means of travel, but much work remains to determine best practices for policy and infrastructure. A city such as Calgary will need a long-term plan to adapt to the design requirements presented by autonomous vehicles, so that future transportation methods can fully utilize maximize potential advantages

Keywords: Autonomous Vehicles, Infrastructure, Drivers, Parking, User(s) Introduction

Research into Autonomous Vehicles (AV’s) has been driven forward by technological advances in areas such as mapping and sensors, communications, as well as machine learning software. [1]

However, the advantages of these technologies are constrained by the ability for both regulators to create policy and governance and by the ability of infrastructure designers to adapt to a shifting vehicle market. [2] The Society of Automobile Engineers (SAE), has defined 6 levels of automation in vehicles, ranging from level 0, where all driving tasks are completed by the driver, to level 5, where the vehicle can perform all tasks independent of a human user. [3]

Currently, there are consumer vehicles on the road that fall under the categories of 0-2 with a variety of tech and motor vehicle companies pursuing various stages of levels 3-5. [4] The problem that remains then is how should civil engineering adapt to these developments? Policy and design will need to adapt to both higher levels of autonomation and changing vehicle user models to ensure safety and efficiency on all roadways. A change in technology will likely drive a shift in vehicle usage that will require a new mindset for developing new land and existing urban spaces. While plenty of opportunities could exist to improve road safety, efficiency and environmental impact, there is a lot to be determined regarding the integration of AV’s with existing road-users, as well as their impact on traffic congestion and infrastructure design.

• How does infrastructure handle both human and computer drivers?
• What impact will Autonomous vehicles have on traffic flows?
• How will changing vehicle usage impact road and corridor design?
• How does AV usage impact urban design and land use requirements?

Combining AV’s and Human Driven Vehicles (HV’s): Autonomous vehicles aren’t going to be adopted overnight, and integration will be a difficult process due to government oversight, user trust, and prohibitive costs. As on-demand services like UBER are actively testing fully automated Vehicles, [5] it is likely that some of the first widespread consumer uses of AV’s will be through shared vehicle usage models, be it ride-hailing, or ride-sharing. At small levels of adoption, self-driving vehicles will be able to integrate relatively seamlessly with regular human drivers. However, as more and more AV’s become available to consumer markets then many of the advertised advantages such as increased road capacity due to smaller lanes and shorter travel distances between cars, or the increased safety benefits due to the removal of human error will require AV’s to be separated from regular human drivers. [6] In a world where 100% adaptation is achieved, this doesn’t prove to be an issue, as human drivers will become exceptions, limited to sports racing, exploration, and to environments where AV software has not yet been trained to handle, (imagine a current AV trying to drive on a highway covered in 2 feet of snow). Additional concerns surround uniformity of infrastructure and road markings. If an AV is dependent on picking up road marking through on-board cameras, then many existing roadways will need updates to ensure consistency and clarity of road markings. [2] To increase a car’s ability to assess road conditions in Calgarian winters, implementing Vehicle to Infrastructure (V2I) communication would allow AV’s to share data with an intersection or laneway. [6] Phasing this technology into new and existing road networks will require large amounts of construction activity to add new ducting, sensors and control systems.

Technology Changing Traffic: Transportation engineers and policy makers will have to study and determine how to adapt new and existing infrastructure to maximize the benefits provided by new technology, installing new, separated laneways on roads and thoroughfares will incur large infrastructure costs to build and design laneways that might ultimately become obsolete as AV adoption becomes much higher. It can be concluded that installing separated AV laneways must be done gradually as market adoption increases, and that there will tradeoffs such as increased congestion for human drivers when AV usage is less than warranted by the new infrastructure, and that sequentially, congestion will lighten for human drivers as more AV’s enter the market. [8] Additionally, while AV’s are integrated with human drivers there will be a need regarding mixed-use traffic intersections where human vehicles and AV’s both interact and behave in different ways.

Additionally, as AV’s have differing sensor and software capabilities, transportation engineers will need to design intersections and traffic light systems to interact with cars using V2I capabilities while maximizing safety and efficiency. [9] Corridor Design for Autonomous Vehicles: As full AV integration is achieved. the requirements for roadways will also change. This is critical to assess now as current infrastructure projects and plans will be built to last for decades to come. A need exists then for the City of Calgary to begin “Future Proofing” [10] both existing highways, and proposed projects. Civil engineers will need to create new standard designs for right of ways, lane use, and lane width to accommodate increased vehicle usage while decreasing congestion due to the advantages of AV’s. Increased vehicle uses, whether more efficient or not, also means that road surfaces will require thicker layers of pavement and more frequent maintenance. Calgary will need to combine these design considerations with their existing complete streets policies. Complete streets, is a movement aiming to better accommodate all road users, not just personal vehicles. These include better infrastructure for transit, bikes, and pedestrians. [11] AV’s and the way that we use them need to be a part of these plans, as narrower laneways, and a nearly eradicated need for street-side parking will create added space for bike paths, bus bays, and boulevards. Another topic that should be considered in corridor design considers public utilities. Most roads provide right of ways to public utilities including water resources, telecommunications, and power.

For example, Calgary requires minimum 2-meter horizontal clearances between water and storm/sanitary pipes sized greater than 100mm. [12] So, while AV’s present the option to create narrower road surfaces, constraints remain on minimum road and corridor right of ways. However, reduced hard travel surfaces does provide an opportunity to promote storm water Low Impact Development practices [13] by enlarging or adding absorbent green boulevard space with bio swales, rain water gardens, and overland flow routes. This is important to ensuring the environmental sustainability of a city and is particularly important to storm water management in Calgary surrounding the Bow River and various flood plains.

Shifting Vehicle Usage and Land Requirements: One of the biggest impacts of AV’s on urban areas will likely be on parking. If a vehicle can drop you off at a business and then search for parking by itself or drive itself to the next user in a vehicle fleet model, then the importance of street side parking to small businesses will be reduced. Additionally, under assumed increases in fleet-style car usage (ride hailing/ride sharing), there will be a generally decreased need for large parking spaces. With the elimination of human drivers both types of services will be able to easily drive from user to user. This will increase road usage but when these vehicles do need places to park, they can be compactly stored in tight depots that require about 25% of the space required by conventional parking lots. [14] One concern that arises from the opportunities of AV’s is an increase of urban sprawl. Most suburban and commuter communities are limited by desirable driving distances from urban centers and office spaces. If personal vehicles become fully automated it means that more people would find communities such as Airdrie, Cochrane, or Okotoks, as more desirable locations to commute into Calgary as driving time would no longer be considered “lost” time as vehicle occupants would be free to sleep, read, work, etc. Some of these issues can be addressed by proper urban planning and zoning to promote residential growth near downtown areas, including promoting urban walkability.