As AV technology evolves, so too will the legal frameworks and safety protocols that govern our roads. Whether these vehicles will be able to adhere to these laws remains to be seen.
Whether we like it or not, self-driving cars, or autonomous vehicles (AVs), are coming our way. And they may or may not be able to recognize us. How reassuring. As we’re forced to acclimate to this emerging technology, we can’t help but wonder how these AVs will adapt to changes in traffic laws. This year’s National Move Over Day provides just the opportunity to examine the likelihood.
National Move Over Day, every third Saturday in October, is a public reminder of fairly new, life-saving laws across all 50 states reminding drivers their responsibility in moving over a lane if there is a vehicle stopped on the side of the road. We need to change behavior on the road and having a National Day of Whatever can often help affirm those changes. If humans are struggling to adapt these new laws, can we expect an AV to be able to also adapt when they struggle with will establish driving behavior? If they can barely stop themselves from hitting a train, are they going to be able to follow a simple lane change?
Understanding the “Move Over” Law
New Jersey’s “Move Over” law, as in other states, requires motorists to change lanes (or slow down if a lane change is unsafe) when approaching emergency vehicles or service vehicles that are stopped on the side of the road with flashing hazard lights. Think police cars, ambulances, fire trucks, tow trucks, highway maintenance vehicles, and regular old citizen vehicles with their hazards on.
These laws have been implemented nationwide because of their positive effect in lowering the injury and death of people on the side of the road. For years, roadside workers and emergency responders are struck by passing vehicles as they perform their duties on the side of the road. Move Over laws are very effective in reducing these incidents, but more awareness is required. Hence National Move Over Day.
Still, as people learn and adapt to these laws, how can we be confident that self-driving vehicles will be able to respond to a sudden lane change?
The Technology Behind Self-Driving Cars
So, how do these self-driving cars actually work? Through a combination of sensors – including cameras, Light Detection and Ranging (LIDAR), radar, and ultrasonic sensors – to detect and interpret their environment. Artificial intelligence algorithms then process this data in order for the AV to make real-time driving decisions, such as acceleration, braking, and lane changes. Despite confidence from the developers of this technology, 93% of Americans have concerns about AVs.
In order for an AV to competently comply with New Jersey’s Move Over law, they would need to:
Recognize the hazard: The car’s sensors must detect and correctly identify the presence of a stationary emergency or service vehicle with flashing lights.
Assess the situation: The system needs to determine whether it is safe to change lanes. If it is, the vehicle must execute a safe lane change. If it’s not, the vehicle must slow down to a reasonable speed as required by law.
Make real-time decisions: The car must adapt to rapidly changing conditions, such as traffic, road markings, and the behavior of other vehicles on the road.
One major difference between AV reaction time and human reaction time is that humans have evolved with this ability while we have to teach technology how to react the way humans do. And that involves trial and error.
Current Capabilities of Self-Driving Cars
AVs are measured on a scale from 0 – 5 by the Society of Automotive Engineers in terms of their relative autonomy. 0 meaning no automation, the driver has 100% manual control over the vehicle; to 5 meaning the vehicle performs all driving tasks autonomously without any human interaction necessary. Level 5 AVs are not available to the public, thank goodness.
The AVs currently on the road are classified as Level 3. This level allows the car to handle most driving tasks, but a human driver is still necessary in case the AV’s system is unable to handle a situation it cannot handle. In a situation where, say, there’s a vehicle on the side of the road and the AV is unable to recognize it, you’d want to have the option and ability to control the car.
Tesla’s Autopilot and Full Self-Driving (FSD) systems, for instance, can perform lane changes automatically. However, the extent in which these systems can handle state-by-state legal requirements like Move Over laws is unconfirmed. Not exactly reassuring, but it’s the current state of the technology.
If self-driving cars are indeed the future, then these vehicles need to be able to adapt on a whim. And it’s not like there will be much warning ahead for a vehicle that’s pulled over. They need to be able to recognize those flashing hazards, whether they’re from an emergency vehicle or a citizen’s, and respond accordingly and to the letter of the law.
Challenges in Complying with the “Move Over” Law
While AVs are not fully there yet, they are definitely improving. In order to improve in the way we need them to, there are a few more hurdles for developers to work into the systems.
Object Recognition: This is the single most important aspect of a self-driving car. AVs need to be able to correctly identify emergency vehicles or road workers, especially in poor weather conditions or low visibility. LIDAR is capable of detecting flashing lights, being capable of recognizing these flashing lights as emergency or service vehicles apart from other road features requires some seriously robust software algorithms.
Lane Changing in Heavy Traffic: AVs can handle simple lane changes and follow basic traffic laws in normal traffic flow. In congested traffic, however, the reliability is a little more questionable. In a situation involving a vehicle pulled on the side of the road, the law states that you have to change lanes or slow down significantly if a lane change is unable to be safely completed. An AV needs to be able to determine on a whim which option is safer and do it without a) bumping into another vehicle or b) slowing down traffic enough to make everything worse.
Adapting to Local Traffic Laws: AVs need to be able to adhere to laws that vary state-by-state. Move Over laws, despite being generally the same, vary by state. The way to do this is with continuous algorithmic updates to the vehicle’s software.
Predicting Human Behavior: Humans aren’t even very good at this. I mean, the FBI is, but day-to-day folks? Not so easy. If people were better at predicting human behavior, you would probably see far less accidents. But, we don’t live in that world. Self-driving cars need to be able to respond to erratic or sudden behavior by other drivers, and more quickly than a human can.
If a driver (or, rider, more accurately) is behind the wheel of an AV that causes an accident, they’re going to need to give a Jersey City car accident lawyer a call.
Regulatory and Ethical Considerations
Should an AV fail to adhere to a Move Over law and cause an injury or, god forbid, death, who is responsible? The car’s manufacturer, the software developer, or the vehicle’s owner? The question of liability is still indeed a question, as regulations are playing catchup to a rapidly evolving AV industry. And regulations are slow to meet the moment in more stable industries.
Regardless of what is considered right by regulatory standards, there are ethical considerations as well. What determines a reasonable speed to slow down? What should the AV prioritize, emergency workers, passengers, or the flow of traffic? These decisions are made in the moment and are even difficult for humans to determine. What can we expect from an AV? And how do developers code that into the algorithm? These are all questions we have to figure out before we let five ton metal boxes drive us on the road.
Self-Driving Cars Are Coming, But They’re Getting Better
Whether we’re thrilled or annoyed, self-driving cars are forcing their way into our lives. But despite that aggressive rollout, our concerns will have an effect on the developing technology. One of those concerns that most people aren’t thinking about is adhering to Move Over laws. These varying legal frameworks are a challenge to software developers but a very worthwhile one. It will be extremely impressive if they bring AVs into a space where they can think faster and more accurately than humans in random, sudden moments. As AV technology evolves, so too will the legal frameworks and safety protocols that govern our roads. Whether these vehicles will be able to adhere to these laws remains to be seen.
Join the conversation!