“Speed, safety and comfort will be the keynote of tomorrow’s highways.”
So said an animated short film called “Magic Highway”, produced by Walt Disney in 1958 (see link above). The film predicted electronic operating devices in cars, teletype machines for communications, and rear-view cameras displayed on a “television screen” on the dash console. It predicted GPS navigation, pre programmed routes and driverless cars where families could relax together while on a trip. The film also foresaw truck-trains that combined “railroad volumes with highway flexibility,” that would deliver goods from remote farms to metropolitan markets in minutes. If 1958 is when the honeymoon with autonomous vehicles began, the hard work getting there started long before that. Possibly the earliest example of an autonomous vehicle was a 1926 Chandler, dubbed The American Wonder. It traveled the streets of New York City controlled remotely by another vehicle following behind. This was possibly the earliest demonstration of platooning, too. Fast-forward to 1991 when the United States Congress passed the ISTEA Transportation Authorization bill. It instructed USDOT to demonstrate an automated vehicle and highway system by 1997. The $90 million program culminated in a demonstration in August 1997 on I-15 in San Diego, California, with 20 automated vehicles, including cars, buses, and trucks driving in “close-headway platoon formation.” There have been dozens of other demonstration projects since then. The first example of a fully functioning, commercialized driverless truck was an autonomous mine haulage truck that began working in Australia in 2008. Since then, advancements in technology enabled functionality like adaptive cruise control, predictive cruise control and servo-controlled steering and braking. When combined with RADAR, LIDAR (LIght Detection And Ranging), high-resolution 360-degree cameras, and artificial intelligence, vehicles became capable of navigating roadways with complex traffic patterns and making driving decisions based on machine learning. Trucks are now literally driving themselves with no human input, but Disney’s vision of the driverless future continues to elude us. We are tantalizingly close, but the challenges that remain on the pathway to commercialization will keep the dream at bay for some time to come. That is why Locomation has focused on a different approach: human-guided autonomous convoys where a human drives the leader truck while the second truck drives itself, following the front truck.
We Still Need Human Drivers
Consider the dynamics of the driving environment, the most complex of all is an urban landscape, with pedestrians, cyclists, crowded streets, tight turns and millions of variables every mile. A less demanding environment is a multi-lane highway where traffic patterns are more predictable and the variables are fewer. From a technology perspective, we can manage commercial vehicles in this environment today, but the AI companies and the trucking companies are still very much on the upside of the learning curve. Eliminating drivers entirely from truck cabs will eventually come, but it’s unrealistic to think we can take drivers from the cab anytime soon. Locomation’s Autonomous Relay Convoy (ARCTM) offers the best of both worlds. The human driver in the leader truck of the convoy sees and understands all the intricacies of the dynamic driving environment while the AI-controlled follower truck literally follows the leader based on its own perception of the leader as well as information exchanged over dedicated short-range inter-vehicle communication. While the follower truck remains aware of all its surroundings and is cognizant of potential threats, it leaves most of the high-level driving decisions to the human in the leader truck and is only concerned with following and mimicking the leader truck. Of course, human drivers do much more than just turn the steering wheel. They are deeply involved in every step of the supply chain workflow. Simply put, the current transportation infrastructure cannot function without some human intervention, whether it is reacting to a highway closure, perhaps an accident where law enforcement personnel are rerouting traffic and trying to clear a jammed highway, or a mechanical problem that requires a pair of hands to solve. The obvious solution is human-guided convoys. The safety and economic benefits are crystal clear. When fully commercialized, fleets will see operating costs drop by about 30%, and convoys will reduce carbon dioxide emissions by over 40 metric tons annually per convoy through an estimated 8% fuel consumption reduction produced by a combination of improvements in aerodynamic drag and precise control of the vehicle by the autonomous driving system. And these human-piloted convoys keep skilled professional drivers employed while easing the pressure on fleets to hire every driver that knocks at its door.
How Close Are We?
Locomation is already running revenue loads for a leading logistics provider, producing revenue for the company while adding thousands of miles and hours of data collection. And we recently completed a pilot project operating on Interstate 84 between Nampa, Idaho and Portland, Oregon, which is arguably a more challenging environment than Interstate 10 between Phoenix and Dallas. That section of I-84 is home to some of the most challenging grades and curves in the national highway system, and it is well known for its gusty and unpredictable winds. While we were not operating fully autonomously over the entire route — the most challenging sections were managed by the human drivers — we had a chance to observe how the system would have performed on its own. We were collecting data all the time and learning from the experience. There are many AI companies now flirting with commercial transportation and developing systems they hope will make it to market. The space is getting crowded, and most are working on the idea that fully autonomous trucks will be on the road very soon. We believe that autonomous trucks are the future, but our pathway to market brings us to commercialization faster, safer, and more profitably for the clients we serve. We are focusing on the parts of full autonomy that can be commercialized successfully today, leaving the parts that require further development and validation for tomorrow. We are pushing the boundaries, for certain, but not to the extent that we jeopardize the future of trucking.