The use of autonomous or self-driving vehicles is rising rapidly across the globe. Experts estimate that the end of this year will see around 26,000 fully self-driving vehicles on roads worldwide, rising to more than 125,000 by 2030. Expand the definition to at least some level of automation. Up to 60% of cars on the road today already fit that bill. These rapid technological advances can quickly become a struggle for transportation agencies. Specific road markings must be captured by autonomous vehicles’ cameras to drive safely. Local and state agencies must ensure the roads are in good enough shape.
One of the three stated priorities in the U.S. Department of Transportation’s comprehensive plan for automated vehicles is to prepare the transportation system. Now is the best time to start preparing on the state and local level. The best place to start is the paint on your roads.
How Autonomous Vehicles Navigate the Road
How autonomous vehicles—from assisted driving to fully automated—navigate the road is complex. They rely on a variety of factors, including:
- Cameras and sensors create a comprehensive, real-time view of the environment around the car.
- High-definition maps rely on detailed imagery to provide information about the exact roads, lane markings, and road signage.
- Computer vision and path-planning algorithms that recognize objects on the road, lane markings, traffic lights, and rules for any given road.
- Localization, using GPS to determine the vehicle’s exact location on a given road down to the inch.
Each system is connected to a database that both feeds and receives information. It’s how real-time updates like road closures, accidents, or weather changes can impact the way the autonomous vehicle drives and navigates.
Precision is key with each of these systems. Technology continually evolves to ensure that vehicles can receive and interpret information from the environment surrounding them as precisely as possible. After all, even a miscalculation of a couple of inches, or missing a single road sign or lane marking, can prove fatal for the vehicle and its occupants.
Top Causes of Accidents for Automated Vehicles Today
According to the National Highway Traffic Safety Administration, self-driving vehicles are more than twice as likely to be reported as part of car accidents as their non-autonomous counterparts. More specifically, driverless vehicles see an average of more than nine crashes per one million vehicle miles. That number is slightly below 4.5 for conventional vehicles over the same span.
To be sure, consider that number with at least some nuance. The NHTSA has strict reporting requirements for any autonomous vehicles involved in a crash. This means that even small accidents with minor damage (and no police involved) are part of that statistic. The continuing evolution of technology will also likely result in that number decreasing soon.
Still, examining the causes of these accidents can lead to crucial insights about making roads safer for autonomous vehicles.
Driver error is no longer a concern in this type of car, but machine error could still cause significant issues. More importantly, the inability to evaluate, analyze, and react to the captured environment quickly, tends to be a recurring issue. According to a 2020 study in a Transportation Research Procedure journal, the types of crash autonomous vehicles are involved in broke down as follows:
- Rear-end accidents accounted for nearly 64% of all crashes
- Side-swipe accidents accounted for more than 20% of all crashes
- Broadside accidents accounted for more than 5% of all crashes
- Collisions with an object accounted for nearly 4% of all crashes
Each type of accident can have a variety of causes. Rear-end accidents, for example, might be attributed to the software being unable to determine in time whether and how fast the vehicle in front of it was hitting the brakes. However, better identification of a potential cause of that slowdown, such as a stop sign, might have prevented the issue.
Similarly, sideswipe accidents may be caused by the software’s inability to recognize a vehicle in another lane. But if the lane markings were difficult to recognize for its sensors, the issue might also be connected to an accidental lane change that could be prevented by better markings.
How Transportation Agencies Are Beginning to Account for Autonomous Vehicles
As you would expect with a new type of vehicle becoming a major factor on the road, state-level governments and departments of transportation have begun to introduce legislation specifically connected to autonomous vehicles. 41 states have enacted related legislation, often leading or in addition to regulations based on the NHTSA’s 2021 Comprehensive Plan on Automated Vehicles.
Pennsylvania offers one example of a state adjusting to this new reality. First assembled in 2016, an AV Task Force continues to spearhead new initiatives, rules, and regulations on anything connected to autonomous vehicles. Representatives from state and local governments, private industry, technology companies, and higher education have been responsible and involved in initiatives like:
- The safe integration of autonomous vehicles in work zones
- Regulations for driverless trucks, which are advancing in their use and potential
- The new guidance, published in 2021, to account for the increasing use of autonomous vehicles when considering new pavement markings
Most other states have begun and advanced in similar efforts. This shows just how important this topic has become for transportation agencies around the United States.
The Critical Nature of Road Markings for Autonomous Vehicles—and Everyone Else
Autonomous vehicles are here and agencies across the country are starting to adjust. While the topic is undoubtedly complex, much comes down to simple truth. Among them: safe road conditions are an absolute must to avoid potentially serious accidents because sensors and cameras can’t read them.
Put differently, agencies must maintain their roads in a way that assists these vehicles in following traffic signage and lines, and remaining within their proper lanes. That starts with painting the lines. Clear markings, in high contrast with the environment around them, and continuous striping are among the best and easiest ways to account for the future of self-driving vehicle traffic.
Of course, clear road markings are not just beneficial for the sensors of autonomous vehicles. Prioritizing this step is vital to ensure that everyone, including but not limited to AVs, remains safe on the road.
