Safe berthings at stops

Why can´t they make a combination of an electric blanket and an automatic toaster? That way, one could get automatically ejected out of bed every morning!

afety at stop sites can be a tricky business with vehicles that do not have human drivers. Although the computer controlled safety systems go a long way to reduce personal hazards, additional measures should be taken. FLYWAY® has 2 alternative systems for safe landings. The FLYWAY "cubicles" are dealt with on this page, while the obstacle detection system is described on another page.	General The FLYWAY cubicles Handling vehicles of different sizes Protecting the berth from snow Operating the doors Handling the safety at the doors Speed and safety at berthings Time to lower the elevator (table)

1. General

he FLYWAY® carriages can berth to load/unload people and cargo in 4 ways: Not using the elevator; berthing horizontally in a protected area Not using the elevator; berthing horizontally in an unprotected area Using the elevator to berth vertically into a FLYWAY® cubicle Using the elevator to berth vertically at an unprotected place, or a place that has other kinds of protection. The word "protected" means that: People, animals and objects are reasonable well protected from being hit by the beamcar´s maneuverings The beamcars are reasonable well protected from rough landings and from bumping into objects. Figure 1:1 gives an example of berthing without using any elevator. The figure shows a protected beamcar platform (above a railway platform), protected insofar as people on the platform are separated from the beamcars by low walls. These walls have automatic doors that work in concert with the vehicle doors. If these walls were removed (and nothing similar, such as hedges or fences were put in their place) the beamcars could change beams more freely, and berth anywhere. But there would be no safety to either the cars or to other objects in the area.

Figure 1:1 When using the elevator, the same type of protection (or even better!) can be obtained with the cubicles, which like wise have automatic doors that work in concert with the vehicle doors. But these cubicles are only practical for passenger service. In most other instances, one would have to take other measures, such as cordoning off cargo areas or using guards. There are of course instances where cubicles for passengers are not desired, such as: there are no space on the ground for such cubicles the operator chooses to not make that investment, but implements other solutions instead the cubicles are not practical, for instance when loading groceries and the like in a booked beamcar the operator prefers to allow the beamcars to berth anywhere along certain beams. SwedeTrack System has given due consideration to all of these instances. We have designed an obstacle detection system for "unprotected" situations. And we have designed the "cubicles". And these are dealth with on this page.

2. The FLYWAY Cubicles

he main problem with automatically controlled cars that are supposed to "land" in the open street, like the FLYWAY system, is that the car could conceivably land upon a sleeping dog or some object lying in the street. The FLYWAY beamcars will have sensors underneath the floor of the cabins, that are contact-sensitive. As soon as something presses on the bottom of the cabin or carriage, the lowering of the cabin stops, and the beamcar checks with its stored information that this is indeed the level where the ground should be. The beamcar´s computer has information stored about how far down the ground is, for each individual stop site in the network. In this way, the lowering of the cabin would be optimally speed-regulated, so it does not land with a "bump!" Now, if the sensor underneath the cabin registers ground contact higher up than expected, the cabin would stop lowering, then alight again. Likewise, if ground contact is not registered where it should be, the cabin will alight. Attendants at the control center would be automatically alerted, and on-board cameras would enable them to check on the situation. They would also be able to manually direct the car to another berth to land. Figure 2:3

One system, with the idea taken from the subway of Singapore, is to use small landing areas that are shielded off from the milling crowd by plexiglass cubicles as shown in figure 2:1. The system is called "Platform Screen Doors" (PSD), and is used in for instance the Metros of Singapore, Paris and Copenhagen to prevent travellers from falling (or jumping) onto the track. In FlyWay®, this idea is used in the form of free-standing cubicles underneath the beams, cubicles that have doors that only slide open when a beamcar is in position on the ground or platform inside the cubicle. This cubicle would have only enough space to harbor the biggest vehicles that will use it. The illustration shows approximate measurements (in meters) for the cubicle intended for a 4-seat beamcar. Figure 2:1 How this plexiglass cubicle could look in a street served by cars that don´t use lifts is shown in figure 2:2 below. This long booth covers the whole stop area and also has to include the lower part of the sloping beams (so that the descending cars won´t hit the walls of the booth).

Figure 2:2

3. Handling vehicles of different sizes

or safety reasons, it should not be possible for persons to squeeze in between the cubicle doors and the vehicle doors. For this reason, stops that are used by vehicles of varying sizes would need cubicles for each width of those vehicles.		The length of the cubicles would be such as to accomodate the longest vehicles for each width, while the width would be suited to the vehicles, as illustrated in the birds-eye view of figure 3:1 below. The various sizes of berths for FlyWay´s cabins have been dealth with on this page.

Figure 3:1

4. Protecting the berth from snow

Figure 4:3 The roof on the cubicle in figure 4:2 gives good protection from rain. If the sidewalls are high enough, it would also give reasonably good protection from snow. But a roofless berth with low walls, as in figure 4:1 gives no protection from any weather. We are not talking about the waiting passengers here; providing waiting rooms from them should not be anything but (maybe) a space problem. But having snow collect within the berth cubicle could be a problem in some climates. It does not work well to have the beamcar land on top of two feet of snow! FLYWAY has 2 alternative solutions for this: Providing heating in the ground to melt the snow as it falls Providing an automatically removable roof over the cubicle. This latter option is readily solved by mounting a roof above the cubicle which is hinged at the sides, as is illustrated in figure 4:2 at right. This roof opens automatically when a beamcar stops above it, and closes again when the beamcar takes off. Figure 4:1 shows a sideview and shortend view with the roof raised. These roofs would be manually closed when it snows, otherwise they would normally be in the open position, even when it rains.

Figure 4:1 Figure 4:2

5. Operating the doors

In the cubicle design described here, we have a double set of doors, where the cubicle doors operate in conjunction with the cabin doors, as is illustrated in figure 5:1 at right. These kind of doors at station platforms are increasingly being used in underground train stations around the world, such as in the new, automatic Metro line in Paris. It is called "Platform Screen Doors" (PSD), as mentioned above. Both sets of doors have to be properly closed before the beamcar can take of, so if both the beamcar doors and the cubicle doors are equipped with the the kind of contact lists described under the next heading, we get double safety against a malfunction in these contact lists. How would the beamcar control the cubicle doors? Well, it could probably be done with some kind of mechanical device. But the solution favored by FLYWAY® would be to use the beamcar cabin´s Bluetooth-based passenger interface for this purpose. The cabin unit (operating as a piconet master) would establish contact with the Bluetooth-unit on the cubicle, and regulate the cubicle door via this link.

Figure 5:1

6. Handling the safety at the doors

egular underground trains occasionally squeeze their passengers in the doors. The doors are of course equipped to handle this automatically. The principle of the most common protective design is illustrated in figure 6:1 at right. The closing edges of the doors have a rubber coating that prevents damage to the squeezed object (or person). Inside this rubber coating are two electrical contact lists, A and B, that are normally separated by an air gap. But the list denoted by B is elastic, and a squeezed object will thus bring it into contact with A, whereupon the train is electronically prevented from starting. The details are better seen in figure 6:2. This system would of course also be implemented with the FLYWAY® cabins.	In the FLYWAY® system; should the doors of either cubicles or beamcabins detect squeezed objects in this manner, the doors would automatically open again. Then, after about 5 seconds, close. Should the squeezed object still be in place after about 4 (or so) closing attempts, an alert would go to the system control center, and appropriate information would be displayed to the passengers in the car, maybe also a recorded message would come over the loudspeakers. Figure 6:1	Figure 6:2

7. Speed and Safety at Berthings

he FLYWAY® system elevators require special attention at the berthing. The safest and easiest way for an approaching vehicle would be to stop above the berth, and then lower itself, as illustrated in A in figure 6. But to save time, the vehicles could also lower themselves as they slow down to a halt, and then raise themselves as they accelerate, as illustrated in B in figure 6. If one choses the B-solution, the obstacle detection radar must scan vertically as well as horizontally, when approaching a stop where the vehicle is about to berth. This is shown in figure 7. The B-solution has one slight drawback, and is therefore not recommended by SwedeTrack: It will put an additional strain on the elevator assembly. Let´s make some simple calculations as to the time-lapse in the 2 alternatives.

Case A: The car is brought to a halt from full speed, lowers the cabin, exchange passengers, lifts the cabin and accelerates to full speed. Case B: The car is brought to a halt from full speed. At the speed of 10.8 km/hour (which corresponds to 3 meters/second), the car lowers its cabin. When the cabin is on the ground, the exchange of passengers take place. The car then lifts the cabin during acceleration. Before the car reaches 10.8 km/hour, the cabin will be all up. The car then goes up to full speed. Considerations for case B: The car must be stationary before the cabin hits the ground. The cabin has to clear the top of the cubicle at the berth, which would be at least 1.5 meters in height at the short ends (where the cabin enters and leaves). Assumptions for both cases: We will calculate the time (T) from where the car starts slowing down, to the point where it´s up to full speed again. The deceleration/acceleration (a) is put to 0.2*g (which is about 2 m/s2). It is not uniform; the "jerk-factor" (j) plays a small role in the equation. It represents the change in the rate of retardation, and is arbitrarily set to 12 m/s3. We will use half that deceleration/acceleration for the lift, i.e. al = 1 m/s2, but use the same jerk-factor as for the beamcar in this example, putting the maximum speed for the lift to vl. "Full speed" (v) is here assumed to be 72 km/hour (= 20 m/s.) Let´s put the height above ground from the bottom of the cabin (d) to 4 meters. We will assume that the car stays one minute at the stop. Regarding the lift, we use the following calculations: d/2 = (vl2)/(2 * a) + (v * a)/(2 * j)

The distance is put to d/2 because we accelerate the lift to maximum speed (= vl), which is reached at mid-point (= 2 meters above the ground), after which we slow down at the same rate, and using the same jerk-factor. Thus, putting the time for the lift to lower itself to the ground (and the same time to go up again) to Tl, we get: Tl = (2/vl)* (d/2) = d/vl = vl/2*a + a/2*j d = 4 m => vl = 1,959 m/s2 => Tl = 2.04 seconds. For case A the calculations are rather straightforward: Retardation and Acceleration: Tr = Ta = v/2*a + a/2*j Tr = 20/2*2 + 2/2*12 = 5 + 0.083 = about 5 seconds. At this low deceleration, the jerk-factor is negligible. So, stopping at a station in case A would take: T = Tr + Tl + 60 sec:s + Tl + Ta = 5 + 2.0 + 60 + 2.0 + 5 = 74 seconds. For case B these times will slightly overlap. The time T3 to decelerate from 20 to 3 m/s is: T3 = 17/(2*2) + 2/(2*12) = 4.25 + 0.083 = 4.33 seconds. So, stopping at a station in case B would take: T = T3 + Tl + 60 sec:s + Tl + T3 = 4.33 + 2.0 + 60 + 2.0 + 4.33 = 72.67 seconds. Suppose we put the speed for starting to lower the lift at 10 m/s instead of 3 m/s. Then, T3 = 10/(2*2) + 2/(2*12) = 2.50 + 0.083 = 2.58 seconds. As can be seen, then, the saving in time is quite negligible.

Figure 31. Figure 32.

A medium-size station for 2 cars in each direction or for 2 cars in one direction could look like figure 31 at left. The sliding doors would only be open when there is a car in position behind them. The roof would be necessary in places where snow can be expected. In other areas, the roof could be dispensed with, and the walls of the cubicle lowered to a hight of about 1.80 to 2 meters (i.e. 6 feet). This would make the station cheaper and less intrusive in the city landscape. It could thus look like in figure 32 (this one with a roof for rainprotection). The sideview picture in figure 33 shows a station for only one beamcar at a time. The station cubicle should not take up more space than an ordinary parking place for motorcars. These small stations could be placed in the streets in crowded downtown areas, one cubicle at each side of the street, and at least one in each direction at every block.

Figure 33. Figure 34 shows what it could look like along a couple of city blocks. The cubicles are the yellow rectangles. It might seem awkward to have them placed in the middle of the street, but one should then remember that the purpose of the beams are to replace practically all private motor vehicles in those streets where the beams are erected, so the traffic in the streets should be just a fraction of what it used to be. An alternative is, of course, to erect the beams along the sidewalks.

Figure 34.

8. Time to lower the elevator (in seconds) as a function of acceleration/deceleration.

t is rather interesting to note how the jerk-factor comes increasingly into play when the acceleration and deceleration increases. The table at right shows the time it takes to lower the cabin, for a range of allowable accelerations/decelerations. The second column shows maximum speed attained (halfway down) in m/sec:s, before the lift decelerates again. The 3:rd column is the sum of columns 4 and 5, and the fifth column contains the term with the jerk-factor, as shown in the calculations to the left. In this table, the jerk-factor is set to 12 m/s.3 and height to 4 meters. For safety reasons, every berth would probably have to be fenced in, using the FLYWAY cubicles, and maybe monitored by surveillance cameras as well, considering that the vehicle has no human driver. But, as stated above, the control staff can always take manual control over any beam vehicle.

Acc/Dec. Max. speed Time (seconds) Term 1 Term 2 1,0 1,959 2,042 1,959 0,083 2,0 2,667 1,500 1,333 0,167 3,0 3,109 1,286 1,036 0,250 4,0 3,389 1,180 0,847 0,333 5,0 3,550 1,127 0,710 0,417 6,0 3,623 1,104 0,604 0,500 7,0 3,630 1,102 0,519 0,583 8,0 3,587 1,115 0,448 0,667 9,0 3,509 1,140 0,390 0,750 10,0 3,407 1,174 0,341 0,833 11,0 3,290 1,216 0,299 0,917 12,0 3,165 1,264 0,264 1,000 13,0 3,037 1,317 0,234 1,083 14,0 2,910 1,375 0,208 1,167 15,0 2,786 1,436 0,186 1,250 16,0 2,667 1,500 0,167 1,333 17,0 2,553 1,567 0,150 1,417 18,0 2,445 1,636 0,136 1,500 19,0 2,344 1,707 0,123 1,583 20,0 2,248 1,779 0,112 1,667