Could a self-driving school bus someday be the norm, leaving the drivers of today to become vehicle monitors who can devote more attention to keeping their riders safe and out of trouble?
Google’s self-driving car project, which has seen success so far on both freeways and city streets, seems to suggest that it’s possible. Self-driving car proponents claim since most accidents are caused by human error — distracted or impaired driving or poor judgment — the technology will reduce traffic congestion due to fewer accidents, resulting in fewer injuries and lives lost and more freedom for those unable to drive.
It will, however, be quite some time and several technological iterations before we get there, and there are still many industry concerns and logistical and liability questions to answer.
More generally, connected vehicle technology (v2v) and automated vehicle technology are currently making great strides for passenger vehicles and trucks, for example, maintaining safe speeds, preventing lane changes that would cause collisions, and sending alerts of unseen pedestrians, but what about for school buses?
In the meantime, school bus manufacturers are keeping pace with offerings such as advanced telematics and air disc brakes.
Self-driving car testing
What if school bus technology advanced to the point where the “driver” was, in fact, no longer human? Google has been testing self-driving cars on the road since 2009, according to the company’s Google+ page, using its self-driving technology, comprised of software and sensors that can detect objects miles away and share environmental data, on Lexus cars on California freeways. An introduction on the Google Self-Driving Car Project Google page says the company developed prototype vehicles over the summer.
In addition to California, Florida; Michigan; Nevada; and Washington, D.C., are allowing these cars on the road for testing purposes, requiring a driver to be present in the vehicle who can take charge if necessary. California is even requiring special permits for driverless cars, of which Google has won 25 of the first 29 being issued, and Audi will be the first vehicle manufacturer to receive one of the permits.
Ralph Knight, administrator on special assignment, fleet manager at Napa (Calif.) Valley Unified School District’s transportation department, says that the district has had a brief phone conversation with Google about school buses: the vehicle design, computer generation and engine function.
In terms of existing passenger vehicles, General Motors (GM) CEO Mary Barra said in September at the ITS World Congress, an annual conference featuring the latest information technology developments, held in Detroit this year, that GM will introduce a 2017 Cadillac model equipped with a feature called Super Cruise, a semi-autonomous technology that drivers can use to put the car into autonomous mode while on the highway. It will maintain a safe speed and use lasers and cameras to ensure the driver doesn’t drift from their lane or get too close to other cars. Additionally, Honda tested an automated vehicle prototype at the ITS World Congress, automotive components manufacturer Denso Corp. demonstrated v2v technology, and Tesla, according to Business Insider, is working on driverless car technology.
Many in the transportation industry whom SBF reached out to for this story say they find the self-driving car concept exciting. The main question asked, though, was exactly how will that technological leap forward safely occur?
Michael Martin, executive director of the National Association for Pupil Transportation (NAPT), says one of many safety issues is the human factor; currently, even self-driving cars need an attentive driver.
“Human error causes 90% of all accidents, so I think the theory is that if you remove the human element, you’ll significantly reduce accidents, and when you use active safety technology and autonomous technology, especially focusing on crash prevention or pre-collision safety, [you’ll] reduce distracted driving,” he says.
However, whether the technology will be widespread enough to see significant benefits will be an issue. For the few initial automated or self-driving cars on the road, there are no other vehicles to communicate with, Martin points out. “There’s also an infrastructure challenge across the board, not just for school transportation; a lot of autonomous technology requires [vehicle-to-infrastructure technology], technology-enabled roadways, and until we have smart roads, having smarter cars is not necessarily going to be a [significant] benefit. You’re going to have to wait for those things to balance out over time.”
Tim Flood, president of the National School Transportation Association (NSTA), along with Martin, says that although self-driving cars may be “a cure for distracted driving, we still need to see what the unintended consequences are.”
Security is a major concern among many who were interviewed, and John Horton, bus driver, Douglas County (Colo.) School District, outlined potential scenarios, speculating that knowledge of a self-driving school bus could potentially attract terrorists, kidnappers, and divorced parents embroiled in custody battles. When he conducted an informal straw poll among his peers and supervisors on the subject, many said having an adult on board for student safety and behavior management would be essential, Horton says.
“The kids [may] say, ‘Hey, we’re riding with no driver, we can do whatever we want,’” he explains. “There will probably be cameras all over the place. However, there’s nothing like having military boots on the ground. Young people on a bus with no driver, that’s almost temptation to do something.”
And that driver should be CDL-qualified, he adds, in case of a maintenance issue, so the vehicle could be driven back to the shop. “You don’t want a driverless car [or school bus] to break down at a point where it would cause a problem, and you need quick access for rescue or evacuation should something go wrong.”
Horton also wonders who police would ticket in the case of an accident or fire, and what resources would be available on the vehicle to notify the authorities when help is needed.
Summing up a consensus among drivers, transportation directors and association presidents who were interviewed, Flood says the technology could help expedite route and stop changes, reduce driver distractions and enable transportation personnel on the bus to “more actively monitor students while facing them instead of occasionally looking through a small oblong mirror. This could help to reduce the occurrences of bullying and students getting out of their seats.”
[PAGEBREAK]Conversely, he notes that while drivers sometimes get distracted for various reasons, “they are also very aware of their surroundings and the neighborhoods they drive in. Will self-driving vehicles adjust to situations on routes? Will they wait for that child that fell running to catch the bus? Losing the human element could be a greater loss than initially thought.”
Horton agrees that losing the human element may take away spontaneous decision-making. He concedes sensors may spot students in the traffic lane and stop the car, but he wonders if they would stop far back enough to provide sufficient clearance or approach school zones and know to slow down.
“Machines do not always make intuitive decisions,” Horton adds.
According to a video on Google’s Self-Driving Car Project page, the sensors can differentiate between a car, cyclist and pedestrian, and can detect how close it is to other objects.
There are also the many different state regulations to be considered. For example, would sensors on the buses be programmed so that in Colorado the service door would close when crossing railroad tracks, and would buses in New Mexico leave the door open to provide an escape route in the event of an accident? Horton asks. According to Google, as it continues to work on the technology, it plans for the cars to comply with the rules of the road in all states.
Les Cross, president of the Ontario School Bus Association and director of business development for Stock Transportation Ltd., agrees the technology could enhance passenger and pedestrian safety. However, he notes, cost is always a big consideration.
“Even though we all think of safety as our primary business, how much can we afford, and how much time goes by before it becomes affordable?” he asks. “That’s always a question for school boards and consumers alike.” (Currently some sources compiled by StateTech, a publication that reviews technology issues faced by government information technology leaders, say the cars, let alone buses, could cost hundreds of thousands of dollars.)
Horton echoes Cross’ concern, asking about the frequency of replacing the backup cameras and other specialized equipment.
By the time self-driving vehicles even get to school transportation, Knight says, they will be more evolved. “It takes a long time for [technology] to trickle down to school buses because you’re usually talking about expensive options, like cars that park themselves, and bus manufacturers know that school districts [have to] buy as cheap as we can,” he explains.
Trish Reed, vice president and general manager of IC Bus, agrees, pointing out that school buses by design are typically a bit later in the cycle to receive and implement state-of-the-art technologies — versus passenger cars — because of the immense focus on safety.
“The safety of our school buses and the precious cargo they transport are the first priority when it comes to designing and building school buses,” she says.
Reed adds the technologies will arrive first in passenger cars, where the volume and scale allows for them to be tested, validated, “and proven again and again before they ever make it to the real world. ... The automotive and passenger car industry is typically the laboratory for innovation and ground-breaking product development,” she says. “Based on the adoption rate of new technologies in the automotive space, you’ll see some of those technologies rolled out in commercial vehicles, and, later, school buses.”
For example, automatic transmissions rolled out decades ago for passenger cars, and are now a standard offering on school buses, she says.
In terms of a massive deployment of self-driving cars, Francine Romine, director of marketing and communications for the University of Michigan Transportation Research Institute (UMTRI), which convenes groups and provides the tools and environment to research technology safely before it takes to the roads, says not to hold your breath.
“There are degrees of automated technology that have to come into play long before you get to what consumers think when you say driverless,” she explains. “Most consumers think there’s a car with no steering wheel [or] gas pedal and it’s just driving down the highway along with me in my car with a steering wheel and a gas pedal. We have a long way to go before we get there.”
However, Romine says, many automakers are working on high degrees of automated technology, such as adaptive cruise control, which is on almost every type of car today.
Automated vehicle technology
A number of the automobile manufacturers have been implementing various levels of autonomy already, such as park assist, adaptive cruise control — the car maintains a safe speed behind the car in front of it, and if that car slows down or stops, the car equipped with the technology will stay a safe distance away — particularly in high end, more expensive vehicles.
Traditional automobile manufacturers are saying they are going to continue to advance autonomous vehicle technology incrementally as they become more comfortable and as people become more comfortable, and that it’s an evolutionary process, says Scott Belcher, president and CEO of ITS America, an organization that advances research, development and deployment of intelligent transportation systems to improve the nation’s surface transportation system.
Other companies, such as Toyota, have said they don’t ever envision what’s called level 4 autonomy, or fully autonomous vehicles, and that a human driver must always be involved, according to Belcher.
On the other side of the debate are Google, Tesla and some other automobile manufacturers that see a more revolutionary transition, going to completely automated vehicles much more rapidly, Belcher says, because they believe that autonomous vehicles are safer than vehicles with human drivers.
“It’s an interesting tension between the legacy automotive industry and the high-tech industry,” he adds. “Tesla is now doing it and [Google] put hundreds of thousands of miles on the road with autonomous vehicles. There are still some challenges, particularly in the urban setting, but I think they feel comfortable that they are addressing those challenges.”
The next challenge will then be policies addressing liability, insurance and privacy. Some parents may think the technology is safer than the driver, while others may think a human driver is necessary. We also may face more complicated discussions about buses than about personal vehicles; an individual makes their own decision about riding in an autonomous vehicle. However, “when you’re talking about an autonomous vehicle with 60 kids in it, there are many more people who have to be a part of the conversation,” Belcher notes.
“Those are tougher questions and those aren’t technical questions,” he adds. “Even if we get to a technical solution, we’ll still have a number of interesting policy questions to address.”
More close at hand is v2v technology, which enables vehicles to communicate with each other and/or the infrastructure by radio signals or Wi-Fi signals about their speed, location and direction 10 times a second to prevent crashes. This technology could, according to the National Highway Traffic Safety Administration (NHTSA), potentially reduce non-impaired crash scenarios by more than 80%.
“That’s bigger than seat belts, electronic stability control and air bags,” Belcher says.
Connected vehicles receive a message if a vehicle is approaching, brakes suddenly or crosses into a lane in front of them and gives the driver an alert in time to respond and avoid an accident. The technology is based on cameras as opposed to radar so the driver doesn’t have to see the obstacle in time to react. For example, if a bus is on a highway and there’s a crash a few cars ahead but the driver can’t see it or if there’s fog, rain or snow, the car will receive an alert, helping it to avoid a pile-up.
Providing an example specific to school buses, Romine says if her car and the school bus that her son rides were equipped with a device sending the signals, she would get an alert while backing her car out of her driveway that the bus was within 500 feet of her car, allowing her to stop and wait for the bus to unload students and continue on its route.
Additionally, vehicle-to-pedestrian technology enables pedestrians’ smart phones or aftermarket devices to send a signal to the driver of the bus that they are stepping out between cars where the driver wouldn’t normally see them. This was demonstrated by Qualcomm and Honda at the ITS World Congress in Detroit in September.
V2v technology and vehicle-to-pedestrian technology will be available through the aftermarket and the cell phone industry so older buses can get access to the technology, Belcher says, and will be deployed in cell phones through a dedicated short-range communication chip in the phone.
The natural progression of vehicle-to-infrastructure technology, and the cornerstone of Google’s self-driving car, is v2v technology, Reed says. “Before you’ll see a driverless truck, you’ll first see a connected convoy of trucks, maybe a driver-operated truck and trailer at the front, a driver-operated truck and trailer in the rear and a driverless, automated truck and trailer in the middle, connected to the other vehicles, knowing when to slow down, when to speed up, when to turn, etc.”
Reed adds that IC Bus is looking into v2v and vehicle-to-infrastructure technologies and has an advanced engineering team working on many of these technologies.
Based on research conducted by UMTRI, NHTSA plans to begin the rulemaking process, Romine says. NHTSA officials say as these technologies evolve, the agency will continue to accelerate its push on innovative and effective solutions to reduce motor vehicle crashes in the U.S., through its research, regulatory and consumer information programs. One of these, its “Significant and Seamless” initiative, promotes sensor-based collision avoidance technology.
Romine says that UMTRI has been the test conductor site for a research project called Safety Pilot Model Deployment, funded by the U.S. Department of Transportation and the largest v2v pilot in the world, with 2,800 vehicles and the world’s largest test bed, an outfitted infrastructure in the northeast quadrant in Ann Arbor.
The pilot, which started in 2011 and concluded in August, supplied data to NHTSA, which evaluated the effectiveness of currently available automated braking systems in avoiding or mitigating crashes and developed test procedures to evaluate the technologies to assess their safety benefits.
The institute recently broke ground on an off-roadway test facility designed specifically for connected and automated vehicles. “It will look like a little downtown area with intersections, obstacles and buildings and is being built specifically to test connected and automated vehicle technologies in a safe environment,” Romine says.
Additionally, one year ago, the University of Michigan created the Mobility Transformation Center (MTC), a public-private research partnership that includes government and academic partners and vehicle manufacturers such as Ford, GM and Toyota. It will focus on v2v and automated technology and research, answering questions such as how the cars will interact with other cars as the technology is being developed, researched and ultimately deployed.
At this point, there are no research projects announced to test the technology on school buses, but the v2v pilot did include school buses that sent a signal with speed, location and direction data that got transmitted as a basic safety message. NHTSA adds, “It is too soon to reach conclusions about the feasibility of producing a vehicle that can safely operate in a fully automated mode in all driving environments and traffic scenarios.” Although public transportation vehicles have been used in its connected vehicle technology tests, the administration says “automated crash avoidance safety systems do not quite exist for these types of vehicles.”
Belcher estimates that v2v and automated technology will roll out in the next three to five years, pointing to work at UMTRI and GM’s announcement. UMTRI currently has 3,000 vehicles deploying the technology in southeast Michigan, including school buses and trucks, and the MTC will increase deployment to 9,000 vehicles, and is working with the Michigan Department of Transportation to eventually move to 20,000 vehicles in the next three to five years, and 120 miles of instrumented roadway. Meanwhile, other states — Texas, Florida, California and Pennsylvania — are also trying to get v2v technology in place as soon as possible because of its life-saving capabilities.
NHTSA issued an advanced notice of proposed rulemaking, plans to issue a rule to require v2v technology in light-duty vehicles such as cars, and is in the process of making a determination about whether to require it in heavy-duty vehicles, possibly this year, Belcher says.
He adds if NHTSA releases a rule in 2016 and it goes through the comment period, the final rule probably wouldn’t go into effect until 2017. Then, it likely wouldn’t apply to new vehicles until 2018 or 2019, and would just start to roll out in new, novel vehicles. However, if GM intends to deploy automated technology in certain new model vehicles beginning in 2017, those models are actually released in 2016, just two years away.
He anticipates several vehicle manufacturers having similar intentions, possibly developing aftermarket devices so drivers can bring them into their cars, buses and trucks, most likely in the next three to five years, and before the rulemaking is completed.
Manufacturers keeping pace
Manufacturers such as Blue Bird Corp., Thomas Built Buses and IC Bus are already offering some automated vehicle features on their buses, such as advanced telematics and air disc brakes.
Kirk Lacko, senior product marketing manager for Blue Bird, says that the manufacturer focuses on active safety technology, with features such as air disc brakes, which help buses stop faster, to potentially prevent accidents.
Blue Bird just launched air disc brake systems on all its 2014 Vision and All American models and is looking into systems related to collision mitigation and electronic stability control.
Another benefit, Lacko says, is there’s less maintenance, resulting in cost savings.
He adds that Blue Bird will roll out some automated advancements in the next model year, making driving even safer and the operator more comfortable, and supports NHTSA’s rulemaking.
IC Bus’ Reed says one vehicle-to-infrastructure connected technology IC Bus has rolled out in commercial trucks that may be added to school buses in the near future is its OnCommand Connection system. The remote diagnostics system takes fault codes from the engine’s computer that have been aggregated into the telematics systems, identifies the severity of the code and provides a streamlined solution to the driver and the fleet maintenance manager. If it’s a minor issue, it would provide a heads-up and recommend the driver take care of it during the next scheduled maintenance appointment. For more severe issues, it would alert the driver and maintenance manager to get the issue resolved as soon as possible and would provide location information for the closest dealer, ensuring that dealer has the necessary part, etc.
While driverless Thomas Built buses are not on the immediate horizon, Ken Hedgecock, vice president of sales, marketing and service at Thomas Built Buses, says the manufacturer does work to incorporate state-of-the-art technology into its buses that make them safer, more durable and more efficient for drivers and students.
“With features such as advanced telematics and student monitoring, Thomas Built Buses continues to stay at the forefront of innovation in the school bus industry,” Hedgecock adds. “We look forward to what the future may bring and to future innovations that will make our buses even better for generations to come.”
Insurance survey: Parents wary of children traveling in self-driving cars
In a recent survey conducted by Insurance.com, a car insurance comparison-shopping website, more than three-quarters of 2,000 licensed drivers surveyed said they would be very likely to buy or at least consider buying a car with autonomous capabilities. When the possibility of much cheaper car insurance as a result of improved safety was introduced, consideration rose to 86%.
However, more than three-quarters of respondents said they wouldn’t trust a driverless car to take their children to school. “Part of it is a natural reaction that you certainly want more precautions taken [with] your child than you normally take for yourself,” Des Toups, managing editor of Insurance.com, says, adding that he is strictly an observer and has no independent knowledge of the production plans of any autonomous vehicles.
Additionally, the wording of the question may have left the type of vehicle open to interpretation. The question, “Would you allow an autonomous or self-driving car to take your kid to school?” may have elicited a different response than using the word “bus,” Toups says. The survey defined “self-driving car” to respondents as “a car that could operate at least part of the time without driver input.”
“Right now there’s no clear universal definition for an autonomous car,” he says. “Basically, any time a computer does something for you — anti-lock brakes and electronic stability control — that’s autonomous technology. We were trying to make sure we distinguished between those types of autonomous technologies and cars that could actually pilot themselves for some period of time.”
Google’s driverless cars aim to make travel safer, faster
Helping improve mobility for everyone, particularly those who cannot drive, is Google’s aim in its Self-Driving Car Project.
According to a post on Google’s project page on May 27 by Chris Urmson, director of the Self-Driving Car Project, self-driving cars will cut down on drive times, allow seniors to keep their freedom, and eliminate drunk and distracted driving.
Part of the project entailed building prototype vehicles designed to operate safely and autonomously without requiring human intervention, relying on its software and sensors, which eliminate blind spots next to the vehicle and can detect objects up to 200 yards in all directions, to help protect against accidents such as those caused by red-light runners, according to the website. Additionally, the speed of the first vehicles is capped at 25 mph.
Google enlisted drivers to work with early-stage prototypes on the company’s test track, taking them through various tests and terrains, and simulated a busy street environment, with “traffic lights, construction zones and wobbling cyclists.”
The company says on its web pages that although its drivers can safely test a vehicle that doesn’t have a steering wheel, it has fit a temporary steering wheel and set of manual controls into each vehicle to comply with California law, which requires any vehicles still being tested to have manual controls. It plans to remove the controls after the prototypes are finished being tested and permitted.
Google plans to run a small pilot program in California in the next couple of years, according to the post.