# Pony Express

## Power loops and Server Sky computation for personal transportation

Note: The following is a work in progress. Please do not assume any of the ideas are complete, or the numbers are accurate. It is intended to demonstrate one application for loop power storage and server sky computation. Thanks to Steve Davis (pilot, toy designer, driving daredevil, and Very Smart Guy) for inspiring the idea and providing valuable feedback.

A 1400 kg Prius accelerates to 100 km/hr ( 27.8 m/s ) in 10 seconds, an average acceleration of 2.8 m/s2, or 0.28 gees. The coefficient of drag is 0.26, the frontal area is 2.16 m2, and the rolling resistance drag is around 0.054 m/s2. The acceleration profile is approximately:

 time speed accel drag power distance energy sec m/s m/s2 || m/s2 kW meters Wh 0 0.0 0.0 0.1 0 0 0 1 4.0 4.0 0.1 23 2 3 2 8.0 4.0 0.1 46 8 13 3 11.6 3.6 0.1 60 18 28 4 14.7 3.1 0.1 66 31 45 5 17.0 2.7 0.1 67 47 64 6 19.2 2.2 0.1 62 65 81 7 21.5 2.3 0.2 75 85 101 8 23.7 2.2 0.2 80 108 122 9 25.9 2.2 0.2 87 133 145 10 27.8 1.9 0.2 82 160 169 11 27.8 0.0 0.2 8 188 171

American drivers need maximum performance at freeway entrance ramps and merge lanes - in dense traffic, automobiles must accelerate to freeway speed and move over into a gap between other cars. Traffic slowdowns also occur at merges. The acceleration requirements determine engine sizes, which determines fuel economy. Making the cars smaller makes them less safe relative to other cars, and making them less agile makes merges more dangerous, so increasing fuel economy may come at the expense of safety.

Modified personal vehicles, plus power loop storage, plus increased global computation, offers a way out of these tradeoffs. Increased energy efficency, safety, economy, and comfort are all possible if we reconsider how we distribute transportation tasks.

## Get a Horse

Horses are less powerful than electric and gasoline engines, but they are smarter. "The horse knows the way" has carried many a drunk home, while drunks and automobiles are a lethal combination. The Pony Express distributed frequent stations across the American west, where riders could exchange mounts and continue their gallop to the next station. Stations were spaced more closely in difficult terrain, because the horses wore out faster. Horses are not long distance creatures - over many hours, a human can outrun them (which is how we captured the first ones).

Some railroad lines add extra engines to freight trains to speed their passage over mountain passes. Like the Pony Express, motive power is modular, not integral to the payload being transported. The extra engines typically have a human crew, but this is a labor union requirement, not a technical or safety requirement. In an age of computers, humans add expense, and in an age of distractions and poor training they often reduce safety.

If we examine an automobile as a component system, we learn that the components may be separable, as they are for these other transportation methods. An automobile has a wheeled carriage, a motor, an energy store, a pilot/driver, and a payload (often just the driver). What if we recombine these components, while adding far more global knowledge, computation, and energy storage?

In an automobile, the navigation needs the most frequent attention. Life-and-death decisions are made many times per minute, in the face of distractions, impairments, and incomplete knowledge. When you think about it, it is surprising that the carnage on the roads is as low as it is, given the chaos and idiocy we encounter out there. While computers often add to driver distraction, they can lead to increased awareness and better decisions. If we add global situational awareness, the computers are probably capable of making quicker and better decisions than human drivers, perhaps acting like a faster version of a self-preservation-conscious horse.

Imagine that we divide automobiles into carriages (perhaps with a low-power electric parking-lot motor) and "horses" - external robotic power-plant and battery combinations that pull the carriages.

Carriages can range from small to large, simple to luxurious. They will be the personalized, individually-owned portions of the system, with the navigation/trip planning computers and stereo systems and safety seats and junk in the back seat. Since they do not contain big road motors, they will not need frequent maintenance and will last much longer. Some carriages may be mere tubing and canvas, like bicycle trailers, weigh less than 100 kilograms, and cost a few hundred dollars. Some may even be pedal-powered for short distances. Some carriages may resemble the passenger compartments of stretch limousines. But even a stretch limo carriage is likely to be less expensive than a luxury car with a powerful engine. Since these vehicles lack motors, they can be parked in smaller spaces, and more can be rolled onto long-distance transports such as car trains.

The "horses" will be small and dense autonomous robotic tractor vehicles, capable of pulling loads much larger than themselves for short distances. A carriage would typically be pulled by two horses, a truck by more. They would come in many sizes, from bicycle pullers to truck-triple-trailer pullers. Horses would probably be owned by energy companies, and rented for 30 minutes until their batteries were exhausted, "replaced in mid flight" by other fully-charged horses, with attach/detach operations in a second or two. Unlike its animal prototype, a motorized horse would be on the road perhaps 12 hours a day, and might need maintenance once a month, major replacements once a year. In cities, the horses will be small and slow, making many quick trips swapping from carriage to carriage. On freeways, the horses will be faster, perhaps more massive. The horses will be fully interchangeable for given classes of carriages, and given the wide range of power that a small electric motor can efficiently provide, a horse might be hauling a smart-car-sized carriage one minute, and part of a 6 horse team hauling a stretch limo the next.

A trip to the store may involve 5% of the charge of one horse. A trip across the U.S. may involve 200 horses, exchanging horses frequently. But unlike a car motor, the horses spend far less time sitting idle. While you spend the night in the motel, the horses that got you there are busy serving other customers, or are traveling to and from centralized recharging stations situated near major power lines. The horse renting companies will strive for maximum utilization, and compete on price and availability with other companies. You may not even be aware which company is providing the horse you are currently using, or how many pennies you are paying to use it. Your decision may be as simple as choosing your horse subscription service, much as you choose your cell phone plan.

Because the horses are not permanently attached to the carriages, they can detach and do appropriate fast maneuvers to get out of the way during accidents. With sufficient situational awareness, two carriages approaching a head-on accident can send their horses away from the crash zone, deploy drag anchors, and slow their relatively light masses at the fastest safe rate, perhaps even spinning 180 degrees so the passengers are thrust into their seat cushions instead of their dashboards. Colliding horses will be expensive, but not as expensive as gravely injured passengers.

Since horses will be small and dense, they may deploy outside of regular traffic on their own narrow roadways, or be hauled around on trucks, perhaps counterflow to the morning commute. They will be fairly dense on the roads, perhaps traveling and waiting for rental in closely-bunched packs. But most will be congregated at fast-charging stations, perhaps drawing on the grid, but in the longer term drawing on power loop storage.

On long steep hills, such as the approaches to mountain passes, horses may pull carriages up the hill, then immediately move over to the downhill lanes, attaching to downhill carriages and charging from regenerative braking.

MORE LATER.

## Communication, Computation, and Control

The horses will have goodly amounts of onboard computers, but the vast majority of the system intelligence will be provided by large data centers, such as server sky arrays. Computers may be stupider than an attentive human being, but they can process vastly more wide-region data. If the horses are autonomous, belong to many different competing companies, and communicate destination information privately with carriages, then the travel routes of carriages and their passengers will be somewhat shielded from "big brother's" surveillance.

MORE LATER

E-cash and electronic payment will be a big part of this system. There will be high demand for horses and power at peak traffic times, and real-time prices will adjust accordingly. Frugal travelers will schedule their travel times and choose their routes to minimize expense - the overall system effect will be to reduce peaks and spread the load. Fully automated carriage/horse systems, and high-bandwidth-anywhere communication, will permit some travelers to let the vehicles do the driving (perhaps with stationary humans in call centers performing safety oversight), so they can work while they commute.

MORE LATER

## Freeway merges

Consider the freeway entrance and lane merge. Busy freeways have metered ramps, slowing the rate at which vehicles are allowed onto the freeway, blindly matching the rate to a predicted density of gaps in the merging lanes. What if the process was smarter than that? If gaps were artificially created in the merging lane, speeding up some vehicles and slowing others with external forces, blocking or redirecting inappropriate lane changes, and maintaining safe spacings in the region of the entry ramp, then an incoming car could be sure of finding a gap to merge into. If the entry ramp itself provided boost power to incoming vehicles, regulating speed for precision gap rendezvous, then less vehicle thrust would be needed to safely merge.

While a Prius needs about 170 watt-hours to accelerate to freeway speed, a big articulated truch needs far more. A four trailer Australian "road train" may weigh 165,000 kg, 230 times the mass of our example Prius, while more modest American and European trucks may weigh 66,000 and 45,000 kg. These big trucks have skilled drivers, but the cars they are merging into may react inappropriately or even dangerously. If a double-trailer truck requires a 80 meter gap in traffic, this can be accomplished by accelerating vehicles in front of the intended gap by 5 km/hr, and decelerating vehicles behind the gap by 5 km/hr, for 30 seconds, as traffic in the merging lane approaches the entry ramp. A 60,000 kg truck could be accelerated to 100 km/hr with 7 kWhrs of kinetic energy, perhaps supplied by some kind of electromagnetic accelerator under the entrance ramp. While it would be expensive (and perhaps unhealthy) to supply an entire freeway system with electromagnetic traction, a few dozen entry ramps, each with a few hundred meters of associated traction speed-up/slow-down roadways, would be far less expensive than increasing the lane capacity of the system.

The energy store for doing this could be a small (100MJ) power loop, charging and depleting many times per hour. It might be partly recharged by regenerative braking from decelerating trucks on exit ramps.

MORE LATER