Paul E. Sandin - Robot Mechanisms and Mechanical Devices Illustrated (779750), страница 34
Текст из файла (страница 34)
It became apparent (though it is unclear if it was theRomans who figured this out) that this caused problems when trying toturn. One or the other set of wheels would skid. The simplest method forfixing this problem was to mount the front set of wheels on each end ofan axle that could swivel in the middle (Figure 6-1). A tongue wasattached to the axle and stuck out from the front of the vehicle, which inturn was attached to a horse.
Pulling on the tongue aligned the frontwheels with the turn. The back wheels followed. This method workedwell and, indeed, still does for four wheeled horse drawn buggies andcarriages.Figure 6-1wheelsPivot mounted front189190Chapter 6Steering HistoryIn the early 1800s, with the advent of steam engines (and, later, electric motors, gas engines, and diesel engines) this steering method beganto show its problems. Vehicles were hard to control at speeds much fasterthan a few meters per second.
The axle and tongue took up a lot of roomswinging back and forth under the front of the vehicle. An attemptaround this problem was to make the axle long enough so that the frontwheels didn’t hit the cart’s sides when turning, but it was not very convenient having the front wheels wider than the rest of the vehicle.The first effective fix was to mount the two front wheels on a mechanism that allowed each wheel to swivel closer to its own center.
Thissaved space and was easier to control and it appeared to work well. In1816, George Lankensperger realized that when turning a corner withthe wheels mounted using that geometry the inside wheel swept a different curve than the outside one, and that there needed to be some othermechanical linkage that would allow this variation in alignment. Heteamed with Rudolph Ackerman, whose name is now synonymous withthis type of steering geometry. Although Ackerman steering is used onalmost every human controlled vehicle designed for use on roads, it isactually not well suited for high mobility vehicles controlled by computers, but it feels right to a human and works very well at higher speeds.
Itturns out there are many other methods for turning corners, some intuitive, some very complex and unintuitive.STEERING BASICSWhen a vehicle is going straight the wheels or tracks all point in thesame direction and rotate at the same speed, but only if they are all thesame diameter. Turning requires some change in this system.
A twowheeled bicycle (Figure 6-2) shows the most intuitive mechanism forperforming this change. Turn the front wheel to a new heading and itrolls in that direction. The back wheel simply follows. Straighten out thefront wheel, and the bicycle goes straight again.Close observation of a tricycle’s two rear wheels demonstratesanother important fact when turning a corner: the wheel on the inside ofthe corner rotates slower than the outside wheel, since the inside wheel isgoing around a smaller circle in the same amount of time. This importantdetail, shown in Figure 6-3, occurs on all wheeled and tracked vehicles.If the vehicle’s wheels are inline, there must be some way to allow thewheels to point in different directions.
If there are wheels on either side,they must be able to rotate at different speeds. Any deviation from thisChapter 6Steering History191Figure 6-2Bicycle steeringFigure 6-3Tricycle steering192Chapter 6Steering Historyand some part of the drive train in contact with the ground will have toslide or skid.Driving straight in one direction requires at least one single directionactuator. A wind-up toy is a good demonstration of this ultra-simpledrive system. Driving straight in both directions requires at least one bidirectional actuator or two single-direction actuators. One of those singledirection actuators can power either a steering mechanism or a seconddrive motor.
Add one more simple single-direction motor to the wind-uptoy, and it can turn to go in any new direction. This shows that the leastnumber of actuators required to travel in any direction is two, and bothcan be single-direction motors.In practice, this turns out to be quite limiting, at least partly because itis tricky to turn in place with only two single direction actuators, butmostly because there aren’t enough drive and steer options to pick fromto get out of a tight spot. Let’s investigate the many varieties of steeringcommonly used in wheeled and tracked robots.The simplest statically stable vehicle has either three wheels or twotracks, and the simplest power system to drive and steer uses only twosingle-direction motors.
It turns out that there are only two ways to steerthese very simple vehicles:1. Two single-direction motors powering a combined drive/steer wheelor combined drive/steer track with some other passive wheels ortracks2. Two single-direction motors, each driving a track or wheel (the thirdwheel on the wheeled layout is a passive swivel caster)The simplest version of the first steering geometry is a single-wheeldrive/steer module mounted on a robot with two fixed wheels. The common tricycle uses this exact layout, but so do some automatic guidedvehicles (AGVs) used in automated warehouses. Mobility is limitedbecause there is only one wheel providing the motive force, while dragging two passive wheels.
This layout works well for the AGV applicationbecause the warehouse’s floor is flat and clean and the aisles aredesigned for this type of vehicle. In an AGV, the drive/steer module usually has a bi-directional steering motor to remove the need to turn thedrive wheel past 180° but single direction steer motors are possible.There are many versions of AGVs—the most complicated types havefour drive/steer modules. These vehicles can steer with, what effectivelyamounts to, any common steering geometry; translate in any directionwithout rotating (commonly called “crabbing”), pseudo-Ackerman steer,turn about any point, or rotate in place with no skidding.
Wheel modulesChapter 6Steering HistoryFigure 6-4trackfor AGVs are available independently, and come in several sizes rangingfrom about 30 cm tall to nearly a meter tall.The second two-single-direction motor steering layout has been successfully tried in research robots and toys, but it doesn’t provide enoughoptions for a vehicle moving around in anything but benign environments.
It can be used on tracked vehicles, but without being able to drivethe tracks backwards, the robot can not turn in place and must turn aboutone track. Figure 6-4 shows this limitation in turning. This may beacceptable for some applications, and the simplicity of single directionelectronic motor-driver may make up for the loss of mobility. Thebiggest advantage of both of these drive/steering systems is extreme simplicity, something not to be taken lightly.The Next Step UpThe next most effective steering method is to have one of the actuatorsbi-directional, and, better than that, to have both bi-directional.
The RugWarrior educational robot uses two bi-directional motors—one at eachwheel. This steering geometry (Figure 6-5a, 6-5b) is called differentialsteering. Varying the relative speed, between the two wheels turns therobot. On some ultra-simple robots, like the Rug Warrior, the thirdwheel does not even swivel, it simply rolls passively on a fixed axle andskids when the robot makes a turn.
Virtually all modern two-tracked193Turning about one194Chapter 6Figure 6-5asteeringSteering HistoryDifferentialFigure 6-5bvehicles use this method to steer, while older tracked vehicles wouldbrake a track on one side, slowing down only that track, which turnedthe vehicle.As discussed in the chapter on wheeled vehicles, this is also the steering method used on some four-wheel loaders like the well-knownBobcat. One motor drives the two wheels on one side of the vehicle, theother drives the two wheels on the other side.
This steering method is soeffective and robust that it is used on a large percentage of four-, six-, andeven eight-wheeled robots, and nearly all modern tracked vehicleswhether autonomous or not. This steering method produces a lot of skid-Chapter 6Steering History195ding of the wheels or tracks. This is where the name “skid steer” comesfrom.The fact that the wheels or tracks skid means this system is wastingenergy wearing off the tires or track pads, and this makes skid steering aninefficient design.
Placing the wheels close together or making the tracksshorter reduces this skidding at the cost of fore/aft stability. Six-wheeledskid-steering vehicles can place the center set of wheels slightly belowthe front and back set, reducing skidding at the cost of adding wobbling.Several all-terrain vehicle manufacturers have made six-wheeled vehicles with this very slight offset, and the concept can be applied to indoorhard-surface robots also. Eight-wheeled robots can benefit from lowering the center two sets of wheels, reducing wobbling somewhat.The single wheel drive/steer module discussed earlier and shown on atricycle in Figure 6-6 can be applied to many layouts, and is, in general,an effective mechanism.
One drawback is some inherent complexitywith powering the wheel through the turning mechanism. This is usuallyaccomplished by putting the drive motor, with a gearbox, inside thewheel. Using this layout, the power to the drive motor is only a couplewires and signal lines from whatever sensors are in the drive wheel.These wires must go through the steering mechanism, which is easierthan passing power mechanically through this joint.
In some motor-inwheel layouts, particularly the syncro-drive discussed next, the steeringFigure 6-6on tricycleDrive/steer module196Chapter 6Figure 6-7Steering HistorySynchronous drivemechanism must be able to rotate the drive wheel in either direction asmuch as is needed. This requires an electrical slip ring in the steeringjoint. Slip rings, also called rotary joints, are manufactured in both standard sizes or custom layouts.One type of mechanical solution to the problem of powering thewheel in a drive/steer module has been done with great success on several sophisticated research robots and is commonly called a syncrodrive. A syncro-drive (Figure 6-7) normally uses three or four wheels.All are driven and steered in unison, synchronously.
This allows fullyholonomic steering (the ability to head in any direction without firstrequiring moving forward). As can be seen in the sketch, the drivemotor is directly above the wheel. An axle goes down through the center of the steering shaft and is coupled to the wheel through a right anglegearbox.This layout is probably the best to use if relying heavily on dead reckoning because it produces little rotational error. Although the dominantdead-reckoning error is usually produced by things in the environment,this system theoretically has the least internal error.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
