Autonomous vehicles have captured the public’s imagination as a self-contained solution to improving transportation. The reality is much more complicated. Below we provide answer to some of the common questions surrounding connected automated vehicles and the role infrastructure can play in creating safe navigation of roadways.
Vehicles with different levels of automated features such as lane keep assist and adaptive cruise control are already on our roads today. These are called automated driving systems (ADS) or advanced driver assistance systems (ADAS),and contribute to safety for drivers and passengers in modern vehicles. The heavy vehicle freight industry has standardized certain driver assistance technologies to help improve the safety and efficiency of truck navigation. There are also ongoing regulatory and research efforts to standardize certain automated driving features for passenger cars (PDF, 2.38 MB).
For instance, the Insurance Institute for Highway Safety (IIHS) studied the impact of lane departure warning and blind spot detection on vehicle safety. The study indicated that these ADS/ADAS features help reduce both the frequency and severity of crashes—lowering rates of single-vehicle, sideswipe and head-on crashes of all severities by 11% and lowering the injury rates caused by these types of crashes by 21%.
Vehicles with higher levels of automation that don’t require a driver to be present in the vehicle and operate within a limited range, such as “robo-taxis” or “autonomous shuttles,” are currently being tested in on-road pilots and may be deployed as soon as 2020-2022.
An article by McKinsey & Company (May 2017) suggests that the introduction of a fully autonomous vehicle for consumers (SAE Level 4/5) is probably over a decade away. Although fully autonomous vehicles may take over a decade to appear on the roads in large numbers, modern ADAS-equipped vehicles and the deployment of robo-taxis and autonomous shuttles require roadway owners to ensure that their road infrastructure is optimized for human drivers, vehicles equipped with automated features and fully connected and autonomous vehicles.
Yes, automated and autonomous vehicles will not be able to navigate the roads without the aid of infrastructure. To achieve the goals of road safety and efficiency, we need to understand how we can use infrastructure to optimize awareness and communication between machines, between humans, and between humans and machines.
The machine vision, radar, LiDAR and other systems that automated vehicles use to navigate the roads and respond to their environments rely on consistent infrastructure, including pavement markings and roadway guide signs, to provide situational awareness and help ensure appropriate responses to their surroundings.
It’s critical to keep infrastructure up to date and to optimize roadways for the machine vision systems used in automated vehicle technologies. According to industry executives, “Shoddy infrastructure has become a roadblock to the development of self-driving cars, vexing engineers and adding time and cost. Poor markings and uneven signage on the 3 million miles of paved roads in the United States are forcing automakers to develop more sophisticated sensors and maps to compensate.”
Additionally, human drivers, motorcyclists, bicyclists and pedestrians will be sharing the roads with automated vehicles for the foreseeable future. This makes it critical to optimize infrastructure for human drivers, as well as automated vehicles.
As vehicles with varying levels of automated features become more common, and we get closer to a future of fully Connected and Autonomous Vehicles, it’s critical to ensure your roads are prepared. To operate safely and efficiently, automated and autonomous vehicles need consistent, high-performance lane markings that provide reliable delineation and retroreflectivity in a range of conditions—including nighttime driving and in dry and wet weather conditions.
Market research predicts a gradual increase of ADS-equipped vehicles on our roadways. Our roads need to be designed to meet the requirements of these advanced driving systems while providing effective communication for human drivers. Deploying separate sets of infrastructure solutions for each user would be costly. However, these solutions don’t have to be mutually exclusive. Roadway owners and operators should consider road infrastructure solutions that enhance safety for both humans and vehicles with varying levels of automation.
Lane markings that help improve the performance of driver assist systems and help improve visibility for human drivers include high-contrast, multi-spectral, day and night visibility and all-weather performance features. Retroreflective roadway signs help improve nighttime visibility and readability for human drivers. Retroreflectivity can also help vehicles with driver assist systems detect and interpret road signs. The durability of lane markings and traffic signage also has a direct impact on machine vision’s ability to consistently detect the infrastructure and efficiently interpret the messages.
Currently, roadway signage is optimized for human vision. This requires ADS/ADAS to employ sophisticated artificial intelligence methods for signage detection and interpretation. These methods help the vehicle interpret surroundings and make decisions based on a “confidence level.” However, they’re not able to adjust to new or missing information as well as humans. Degraded and obscured road signs also present challenges to ADS/ADAS. Roads need to be prepared for ADS/ADAS sensor technologies with road signs optimized for different sensor modalities.
Inconsistent, non-uniform lane markings also present challenges to ADS/ADAS. The sensor systems on ADS/ADAS look for consistent lane marking edges, and can be confused by other longitudinal markings on the road surface such as cracks, fills and road imperfections. Inconsistent lane markings make lane keeping more difficult by reducing the safety margin against these stray surface features. Additionally, deteriorated lane markings lead to reduced visibility.
Consistent, high-performance lane markings are one piece of critical infrastructure required to prepare for vehicles with ADS/ADAS systems. Advanced driving systems such as lane departure warning (LDW) and lane keep assist (LKA)are looking for pavement markings that are in good condition, visible in inclement weather conditions and provide sharp contrast between the road and the marking. These features help vehicles with ADS/ADAS perform more safely and efficiently, and provide better detection for both humans and machine vision system. .
The sensor systems on ADS/ADAS-equipped vehicles look for consistent lane marking edges, and can be confused by cracks, fills, roadway imperfections and other longitudinal markings on the road. Over time lane markings wear deteriorate, which can create additional challenges like reduced lane visibility.
High-quality, high-performance, high-contrast pavement markings help mitigate some of these challenges. Additionally, standardization across road surfaces and regions can help accelerate the optimization of this ADS/ADAS technology.
3M is currently working with the automotive industry, including automotive manufacturers, Tier 1 and 2 suppliers and roadway authorities (PDF, 144.59 KB), to understand challenges with sensor systems and develop products and standards that will help improve safe navigation for all levels of automated vehicles by enhancing roadway infrastructure with low-cost road safety improvements that have been optimized for automated and autonomous vehicles.
Automakers, suppliers and sensor manufacturers ensure safe operation of their systems by following a standard of practice known as Functional Safety. The standard provides a framework for the design, validation and risk assessment of automotive safety systems in the event of failure, and is defined under the internationally recognized standard of ISO 26262 ("Road vehicles – Functional safety").
Safety is one of the key issues and priorities of future automobile development. With the trend of increasing technological complexity, software content and mechatronic implementation, there are increasing risks from systematic failures and random hardware failures. ISO 26262 includes guidance to avoid these risks by providing appropriate requirements and processes.
With the development of automated and autonomous vehicle systems, an additional standard known as “Safety of the Intended Functionality” (SOTIF) is being defined under ISO 21488 to ensure operational safety during normal operation, not just in the event of failure. The SOTIF standard includes assessment criteria based on environmental factors such as adverse weather and road conditions. The SOTIF standard together with ISO 26262 is intended to cover the full life cycles of vehicle safety systems.
3M™ believes that infrastructure materials such as roadway signs and pavement markings, as well as work zone devices, play a critical role in maximizing safety. The machine vision and sensing systems on ADS/ADAS and autonomous vehicles rely on infrastructure as inputs to enhance the vehicle’s situational awareness and help it make safe decisions. 3M™ Connected Roads is developing new infrastructure materials specifically optimized for these systems to help improve road safety and to address the functional safety requirements they must meet.
Here are a couple of resources to learn more about ISO 26262:
Yes, advanced driver assist systems such as lane departure warning (LDW) and lane keep assist (LKA) require consistent, high-contrast pavement marking lines for safe and efficient operation. LDW/LKA systems use machine vision and sensor systems that rely on pavement markings for lane guidance. These systems perform more consistently and provide increased safely and efficiency with durable, consistent lane markings.
There are varying methods used by the automotive industry in their LDW/LKA systems. Some are more sensitive when detecting certain road surface imperfections (ex. road scarring, tar seams, etc.), which may cause false positives. Others may be less sensitive to these road surface conditions. Regardless, there are challenges that LDW/LKA systems encounter when interpreting lane markings—including inconsistent, non-uniform markings—that make lane keeping functionality more difficult by reducing the safety margin against other stray surface conditions.
Our new 3M™ Connected Roads Contrast Tape Series 380ADAS has been specifically designed for detection by both cameras and human drivers at night and in rainy or wet conditions. 380ADAS provides high contrast between the road and the marking and has a durable construction that is stable over the life of the product.
We have conducted research at 3M’s Transportation Safety Research Center in Cottage Grove, Minnesota and with auto manufacturer and Tier 1 suppliers, studying how camera systems associated with ADS/ADAS detect pavement markings. To learn more about the importance of contrast in the pavement marking detection, download our whitepaper: Improved Daytime Detection of Pavement Markings with Machine Vision Cameras
*Our new pavement marking, 3M™ Connected Roads Contrast Tape Series 380ADAS, has been evaluated both in our internal lab and test track as well as with our partners in the automotive industry. Continental Automotive Systems Inc., a top Tier 1 automotive supplier of ADAS systems, including cameras systems associated with LDW/LKA, evaluated Contrast Tape Series 380ADAS and stated: “In nighttime rain conditions, 3M 380ADAS enables detection at longer distances than glass beads on white markings.”*
Our work with partners in the automotive industry like Continental, as well as our continued partnership with roadway owners like state DOTs and local roadway authorities, allows 3M™ to “connect the dots” needed to enable the future of CAVs while working towards the goal of improved safety and mobility for the traveling public.
*Based on testing jointly conducted with Continental Automotive Systems Inc. in December 2017 in Brimley, Michigan, USA.
Vision for Safety 2.0 is federal guidance for Automated Driving Systems (ADS) provided by the U.S. Department of Transportation and National Highway Traffic Safety Administration. This guidance calls for industry, state and local governments, safety and mobility advocates and the public to lay the path for the deployment of automated vehicles and technologies.