PDA

View Full Version : Designing a Safety Study




Seeker
08-31-2002, 03:11 PM
Hi everyone !

I'm looking forward to great discussions on this site. It's an excellent idea to have a moderated forum, where you can discuss things in a mature and intelligent manner...as friends.

Now to my question :

Much has been written on TIQ about the safety of the Segway. If you were to design a series of tests which comprise a research study, to evaluate how 'safe' Segways are on sidewalks, what would this research study look like ?

What are some of the factors that you would have to look at, which would enable to come up with a 'global conclusion' about how safe the Segway was on sidewalks, and which areas might need to be addressed, in order to make it more safe to operate ?

Seeker




Casey
08-31-2002, 03:29 PM
Welcome Seeker,

The forum you describe is our goal.

I am finding things slow on all the forums this morning so we might not get much discussion today.

I'm going to have to do some pondering of your question as to designing a series of tests for Segway. It is such a unique device that it presents questions not present for other types of transportation. The DS system probably precludes barrier testing, unless you somehow cause it to run at full speed without a live rider persent. Otherwise the riders reaction to the barrier collision would negate the results.

I have wondered about an event such as a rider who has passed out leaning forward. Seems to me that would circumvent the DS, because there would be no rider response. In that case the Segway would not stop until it crashed into another object. Would that then throw the rider backward and reverse it? If so would it be soon enough to prevent a catastrophe?

Obviously it is as easy to fall off as any other two wheeled vehicle. But in the case of Segway, it should immediately co;me to rest as opposed to a bicycle or motorcysle etc continuing forward uninhibitted until collision with something.

There are other aspects that need real world testing, as well as controlled tests. Maybe Lawrence, who is active right now might give us his ideas on this too.

Frank

n/a
08-31-2002, 03:34 PM
I think there will have to be a number of different kinds of studies conducted in order to do a thorough assessment. Segway's performance under differing conditions need to be assessed eg. on ice, snow, slippery surfaces, crowded streets, narrow sidewalks, slopes etc.

To convince the anti-segway crowd, studies will have to be done on the results of impacting another human with full load at top speed or the effect of 2 segways colliding at top speed. They may have to show what happens when Segway impact "low level" objects such as animals or small children. Stopping distances will have to be measured under various circumstances and compared to stopping distances of pedestrians or those on other types of vehicles.

I would think that these studies may have to be done by independent testers, ie., not connected to Segway.

Assessing in how well Segways mingle with crowds might be necessary, perhaps indoors and out. Safety comparisons will need to be done ie., will children/youth, drunks be safer using Segways then other forms of transportation?

I really dont think all this should be necessary in order to allow Segways on sidewalks but I think it would be prudent of Segway to do as much testing as possible because if they dont, Segways may quickly disappear from the sidewalks after a few highly visible accidents.

ziggystardust_
08-31-2002, 03:55 PM
I think that testing Segway 'safety' is more a test of human nature than of the device itself. And human nature is a tough thing to quantify.

The Segway itself is an inherently safe device given that it is used 'as intended'. but technology aside, there's not much to the device. 2 wheels, a base, a shaft, and a handlebar and open air. It looks like a scooter. How you balance on it and make it 'go' has no bearing on what it will do if it hits someone or if the rider falls off of it or runs into a wall with it, other than perhaps enhanced action of motion since reaction can be instanteously translated to action for forward and back (not left and right) motion.
Cars go throug so much testing because of all the components, all the variables, the high rate of speed

The laws of physics still come into play with the Segway. I think we can all imagine what it would feel like to get hit with an 80 pound device carying a 180 pound person at 12 mph. wouldn't feel too good. You don't really need to test that to know that injuries would occur to both the pedestrian and the rider. try to turn hard left or right while moving forward at a rapid pace and you'll tip over, hit a pothole in the ground and you can be thrown off, get the wheels caught in a tree root and you can face plant, have one wheel go off the sidewalk and you can topple over. Hit another Segway rider and you can both be injured. Lose traction and the Segway loses the ability to dynamically stabilize. But used 'as intended' and within the laws of physics it will, like most other modes of transport, be inherently safe.

The real test is to just see it in real life. How do people use it? How do pedestrians co-exist with it? If it is done in a formalized fashion obviously people are going to be on their best behavior and you will get skewed data. The real test, fortunately or unfortunately, is just to get them out there and see what happens.

The Zigman

yop
09-01-2002, 01:24 PM
I don't see why they can't do safety studies with real live people. Put the test subjects in some extra safety gear. Should be safer by far than, say, tackle football.

yop
09-01-2002, 01:33 PM
Anybody know offhand how much power the motors in the Segway have? Shouldn't it be possible to get a ballpark figure for stopping distance from that number?

n/a
09-01-2002, 03:43 PM
Zigman writes:
quote:But used 'as intended' and within the laws of physics it will, like most other modes of transport, be inherently safe.

Briclin who has done fairly extensive independent testing, believes that Segways are safer than bikes under a number of conditions. Segways are certainly safer at slow speeds.

quote:The real test, fortunately or unfortunately, is just to get them out there and see what happens.

Maybe not. Maybe they Segway LLC will have to establish some new rules of Segway use/etiquette before testing them under "natural" conditions will make any sense. I imagine the initial "tests" will involve designing plans and procedures for introducing Segways in various cities. People will have to be trained not only on how to ride a Segway but how to use them and mingle with pedestrians. Segway LLC does not want any incidents shortly after launching the consumer models because they would risk getting slapped down with a lot of regulations that would squelch sales. The initial high price will surely guarantee that mature, responsible adults will be the first to use them.

yop
09-02-2002, 09:21 AM
The Segway motors have 4 horsepower. That should be enough to give it a stopping distance at least as good as any jogger. I can run faster than 12mph, and I'm a lot weaker than four horses.

4 horsepower seems excessive for acceleration. Four horses to pull a wagon weighing less than 400 pounds? The power must be there to accomodate deceleration.

Casey
09-02-2002, 09:31 AM
An article Charmed posted elsewhere shows I think, Brazilian police using scooters,(not Segway) to patrol a courthouse, and gois on to quote the police spokesman as saying each scooter frees up about 10 officers for other duties.

What I'm thinking concerning Segway's 4 HP is that when the software in modified to allow the 17 MPH top speed, there will be a need for extra power, and acceleration for police use might also benefit from it. Kind of like a police interceptor engine, that is, for the time being, being restricted for normal use.

Frank

ftropea
09-02-2002, 01:26 PM
There was a thread about this on TIQ and Crash Test Dummies. My feeling on saftey studies on the Segway-HT which might be performed similiarly to how they test cars, would require a specially configured "dummy" that was somehow able to mimic a person standing. I think that because a person seated is less rigid than a person standing. When seated in a car, few muscles are at work. So if you crash into something, your limbs are more likely to fly forward and bend whichever way momentum takes them. But on a Segway-HT, where you're standing and inclined to pull back when stopping, the dynamic of you're movement when hitting something have to be different. The very fact that you may be leaning back as you approach something might dramatically change the conditions and perhaps decrease your risk of sustaining injury.

I could be way off on this... but I'm just thinking that you need to carefully design safety test for the Segway because it's unlike anything else out there.

Regards,

Frank A. Tropea
----------------------------------
Ginger : Robotics :: PC : Computers

charmed
09-02-2002, 01:43 PM
The whole idea of safety testing is odd, to me. There aren't tests on bicycles, arguably a much more prone to injury platform.

At least at 12 mph., it is clear to imagine the forces at play, and what a body and a machine will do when striking an immobile or mobile object.

Basically, anything is possible (unlike a particular car at a particular speed hitting a particular wall). Just being a stand-up platform will be a great safety benefit compared to the bicycle, IMO. How to test this with dummies or automotive type tests is unclear. I think feedback from unbiased bicycle riders will be the most compelling information regards safety, that and the eventual data that will come from real-world usage.

I think those involved in the legislative process realize this to some extent.

Systems like braking, turning, acceleration, etc., could and should be tested and the results made public. Perhaps some independent lab already is doing this?

Seg-wager
09-02-2002, 04:07 PM
&

quote:How to test this with dummies or automotive type tests is unclear. I think feedback from unbiased bicycle riders will be the most compelling information regards safety, that and the eventual data that will come from real-world usage.
There was a thread about this on TIQ and Crash Test Dummies. My feeling on saftey studies on the Segway-HT which might be performed similiarly to how they test cars, would require a specially configured "dummy" that was somehow able to mimic a person standing
[/quote]

A pressure sensitive Asimo? J/K ;)

For stopping time & distance and relative impact-- a treadmill with a pressure sensitive flag that pops up unexpectedly in front of the rider.

Casey
09-02-2002, 04:15 PM
quote:To hit someone with it, you'd be better off actually picking it up and heaving it at the victim, since it can be stopped with an outstretched hand.


http://www.theitquestion.com/forum/topic.asp?TOPIC_ID=10732&FORUM
_ID=1&CAT_ID=1&Topic_Title=it%27s+a+whole+lot+of+fun%2E+PostalMag&Forum
_Title=Segway+discussion

Frank

I'll volunteer my outstretched hand for the barrier crash test.;)

n/a
09-02-2002, 06:23 PM
quote:The whole idea of safety testing is odd, to me. There aren't tests on bicycles, arguably a much more prone to injury platform.

Since Segway is seeking special status, is intended for sidewalks, and claims to be safer than bikes etc., I think they should do as much testing as possible to quiet the worriers. Hopefully the results will be convincing.

Brooster
09-02-2002, 06:41 PM
Charmed writes:

"The whole idea of safety testing is odd, to me. There aren't tests on bicycles, arguably a much more prone to injury platform."

That's an excellent point. Were bicycles safety tested when they were first invented who-knows-how-long ago? Heck, I doubt it. But of course back then, our cities probably weren't so congested and they probably weren't such a litigous society.

Clearing the way for Segway in 2002 is made MOST difficult by the fact that it IS 2002. It's in a way ironic that one of the very problems Segway was invented to address--congestion in our cities--is one of the biggest stumbling blocks to its acceptance.



Brooster

charmed
09-02-2002, 07:02 PM
My problem with testing are those that want to know what happens when you ram a 4 yr. old kid with a Segway at 12 1/2 mph. There is no way to quantify that. Or what happens when you bump a pensioner just venturing out onto the sidewalks for the first time since hip surgery. Or what happens when you hit a brick wall full on. Or what happens when you errantly slide off the sidewalk with one wheel.

Crash tests don't apply, just as they don't with bikes, skates, xooters, etc. Yes, people are bound to get hurt in direct proportion to their mistakes, misjudgements, and contact with outside influences.

Speed is already addressed. Stopping distances would be good to know, I suppose, but something tells me Segway is purposely mum on the point because in essence, it is an extension of one's body and mind, and stopping distances differ from person to person, depending on their reflexes, physical condition, etc. With a car these days, the moment you hit the brake peddle and the anti-lock brakes kick in, you will stop in a predetermined length of roadway. With the Segway, with no brake peddle, it becomes more subjective.

I think there plan of introducing them first to people who work on their feet is a good one. So far, except for one small mishap in Atlanta, they have been and continue to be proven safe and sane. This to me is a good strategy for something that doesn't easily lend itself to the kinds of tests we are used to for other vehicles.

I hope this doesn't sound like I oppose safety testing, just that I perceive that there are inherent vagueries with such a product. It's a little like testing roller skates for stopping distance.

Casey
09-02-2002, 07:30 PM
Actually Charmed, according to this some Norwegien shipbrokers managed to pile some of them up after "a long Friday lunch". Maybe Lawrence can check for us on the status of their sobriety at that point.[:0]


quote:Segway
Having successfully embraced fixing ships standing up, Norwegian shipbrokers take things one step further and are issued with Segway Human Transporters, enabling brokers to wiz around the office at speeds of up to 12 mph.

Whilst 50 % of the brokers can't figure out how to use them, the other 50% are involved in a huge pile up in the lobby of The Theatre Cafe after a long Friday lunch. Three brokers are hospitalised.



Frank

n/a
09-02-2002, 07:33 PM
quote:It's a little like testing roller skates for stopping distance.

If somebody launched an effort to get rollerblades accepted on sidewalks, it would be appropriate to test them for stopping distance etc.

I agree with your point though Charmed. Bikes and other non motorized vehicles have proven to cause a lot of injuries. These vehicles dont have to be tested as some expect the Segway to be tested. In a way it seems unfair. But then none of the manufacturers of these vehicles have posed any threat to the automobile industry and none of them launched a nationwide lobbying campaign to get acceptance on sidewalks.

charmed
09-02-2002, 07:56 PM
quote:Originally posted by Lawrence

quote:It's a little like testing roller skates for stopping distance.

If somebody launched an effort to get rollerblades accepted on sidewalks, it would be appropriate to test them for stopping distance etc.

I agree with your point though Charmed. Bikes and other non motorized vehicles have proven to cause a lot of injuries. These vehicles dont have to be tested as some expect the Segway to be tested. In a way it seems unfair. But then none of the manufacturers of these vehicles have posed any threat to the automobile industry and none of them launched a nationwide lobbying campaign to get acceptance on sidewalks.


I don't take issue with you, Lawrence, accept to make sure you do get my point. My point is that Segway IS akin to roller skates, in that it is going to behave differently for different riders. Some riders will have the reflexes and agility to do some amazing things, and others will be more sedate.

Since it is an extension of one's mind and body, one can expect a varied stopping distance.

I am all for any kind of testing that anyone feels Segway ought to endure. I just think there may be reasons behind Segway's chosen method of introduction and legislation, which so far, doesn't seem to include independent lab tests.

Seg-wager
09-02-2002, 08:38 PM
quote:I'll volunteer my outstretched hand for the barrier crash test.

hehheh. ditto if I can try it then myself. ;)

Antagony
09-06-2002, 07:51 PM
quote:I am all for any kind of testing that anyone feels Segway ought to endure. I just think there may be reasons behind Segway's chosen method of introduction and legislation, which so far, doesn't seem to include independent lab tests.

Your faith in Segway LLC is touching Charmed, but possibly misplaced. If I believed that I had an extremely safe vehicle, I would do all sorts of tests and publish the results before there arose concerns about its safety, to prevent rumours from arising, not afterwards. Once safety concerns become an issue, it will be more difficult to stop the rumours from spreading.

charmed
09-06-2002, 08:01 PM
You do realize, Antagony (I LOVE that name) that Segway has undergone thousands of hours of testing in real world situations on the street and sidewalks of our great nation?

95% of all user reports comment on the incredible control the riders had over this new-fangled conveyence.

Like I said earlier, bring on the testing. We don't know that extensive tests aren't happening as we type.

It's an extension of the rider. Safe rider=safe segway, by all accounts so far (except for that rabblerouser who slammed into the chairs).

Seeker
09-07-2002, 12:58 PM
Another thought on the topic of safety. While I was out for a walk yesterday, I thought about how Segways would compare to devices like rollerblades & scooters, in situations where the rider encounters an obstacle, and is forced to swerve.

Segways have large,soft tires, which among other things, means that they should ride roughly as well on grass, as they would on pavement. What this means, is that if you suddenly realized you were heading for a collision with someone on the sidewalk, you would feel comfortable in moving over to the grass, to avoid crashing into them.

With a scooter or rollerblades on the other hand, you would know that if you moved over to the grass, you might lose your balance. So while rollerbladers and scooter riders would likely move to the grass in situations which they deemed 'quite dangerous', these riders might often opt to try to make an evasive maneuver on the sidewalk itself, where they were accustomed to riding, with control.

I think that these considerations, taken together with the
fact that Segways can negotiate turns more easily, would mean that Segways would have a better record of success in avoiding collisions, than other potential sidewalks devices would have.

Seeker

Seeker
09-09-2002, 11:58 PM
I remember seeing a patent which discussed the use of an airbag for the Ibot. This makes me wonder, could a Segway be rigged up to deploy an airbag ? Since the handle (pole) is leaning forward, if you struck a pedestrian ,the pole would need to move in a distance, before the bottom part of the Segway would contact the pedestrian. Would this be enough time for an airbag to go off, in front of the Segway ? Is there a way in which this could be feasible ?

Seeker

baantjer
09-10-2002, 07:48 AM
quote:Originally posted by Seeker



I remember seeing a patent which discussed the use of an airbag for the Ibot. This makes me wonder, could a Segway be rigged up to deploy an airbag ? Since the handle (pole) is leaning forward, if you struck a pedestrian ,the pole would need to move in a distance, before the bottom part of the Segway would contact the pedestrian. Would this be enough time for an airbag to go off, in front of the Segway ? Is there a way in which this could be feasible ?

Seeker

That was for the IBOT in case it toppled over. In the same way as they also filed a patent that allows the seat and the rest of the IBOT to separate so that in case of sideways or forwards fall the impact can be made less.

n/a
09-10-2002, 11:25 AM
I think that these considerations, taken together with the
fact that Segways can negotiate turns more easily, would mean that Segways would have a better record of success in avoiding collisions, than other potential sidewalks devices would have.

Rollerblades and bikes do have one advantage over the twowheeled Segway. That is that when in motion, their "footprint is only a couple of inches wide. That means that if a path is relatively narrow, bikes and rollerblades would be able to get by whereas Segways could not.

I see that I am starting to stray from the topic. So I will have to give myself a little reminder...stay on topic Lawrence!;)

Casey
09-10-2002, 11:33 AM
quote:That means that if a path is relatively narrow, bikes and rollerblades would be able to get by whereas Segways could not.



Maybe somewhat narrower, but if the rider is still on it there would still be a need for a path at least a foot or more wide. Not a huge advantage. And unless the pedestrians squat very low, the bicycle handlebars would hit them.

n/a
09-10-2002, 11:48 AM
quote:Maybe somewhat narrower, but if the rider is still on it there would still be a need for a path at least a foot or more wide. Not a huge advantage. And unless the pedestrians squat very low, the bicycle handlebars would hit them.

I noticed this fact while riding a bike. There was a pool of water and mud in the middle of a path, on each side there was dry ground a couple of inches wide. I was able to avoid the mud puddle by using the narrow path. This would work with all low lying obsticles. A uniwheeled Segway would solve this problem though.

Glider
09-12-2002, 06:01 PM
Thanks Casey for pointing this thread out.
All in all, I must agree with Lawrence that Segway must be held to a higher standard since they are changing the conception of a pedestrian on sidewalks, unlike bikes and rollerblades.
Because of the proximity to children, strollers, and elderly or wheelchair riders, Segway can't be percieved as a threat to their safety.
Thanks to ftropa on the TIQ link to Crash Dummies.
There was a point, and no discussion here or there that I could tell, regarding
quote:I'd also like to see a yaw steering "startled rider" manoeuvre, but it may be difficult to rig.

From further research (here, the injury center, and the TIQ), it appears that this refers to the control bar's throttle which provides left and right steering through clock or counter twisting. If the rider is "startled" and recoils back, then the steering could be affected (clockwise pull = sharp right turn???) and the rider's forward motion could cause them to fall over the fender.
Is this a real world case - are there motorcycle riders out there that can confirm the issue of "startled rider" inducing unwanted accelerations? Does this translate to the Segway?

Finally (sorry), if the rider on a Segway wanted to jerk the throttle back during full speed operation, could he hold on if he shifted his wieght into the turn?

Glider
09-13-2002, 01:48 PM
I can't find anything after further research on this subject.
My own experience riding the Segway did not allow enough time for me to experience this effect, much less the preknowledge to test it.
Anybody?

quote:Originally posted by Glider

Thanks Casey for pointing this thread out.
All in all, I must agree with Lawrence that Segway must be held to a higher standard since they are changing the conception of a pedestrian on sidewalks, unlike bikes and rollerblades.
Because of the proximity to children, strollers, and elderly or wheelchair riders, Segway can't be percieved as a threat to their safety.
Thanks to ftropa on the TIQ link to Crash Dummies.
There was a point, and no discussion here or there that I could tell, regarding
quote:I'd also like to see a yaw steering "startled rider" manoeuvre, but it may be difficult to rig.

From further research (here, the injury center, and the TIQ), it appears that this refers to the control bar's throttle which provides left and right steering through clock or counter twisting. If the rider is "startled" and recoils back, then the steering could be affected (clockwise pull = sharp right turn???) and the rider's forward motion could cause them to fall over the fender.
Is this a real world case - are there motorcycle riders out there that can confirm the issue of "startled rider" inducing unwanted accelerations? Does this translate to the Segway?

Finally (sorry), if the rider on a Segway wanted to jerk the throttle back during full speed operation, could he hold on if he shifted his wieght into the turn?

charmed
09-13-2002, 02:10 PM
Hiya Glider

you ask about motorcyclists' experiences with twist throttles. I have owned several over the years, and never had the problem you seem to fear. I did once have an ATV with a thumb lever throttle that took some getting used to (ie. hit a bump and vroom ahead) but this was easily mastered.

Regarding a sudden full twist of the steering control at speed--I always assumed that the circuitry would impose limits on turning relative to a given speed. In other words, moving slow, faster turning, moving fast, slower turning. Segway, from what I have seen, has arguably the world's most advanced interface for a ground based vehicle. I can't imagine that they would allow the rider to force the unit into a pirouette at 12.5 mph. when it would be so simple to program in proportional limits. More likely that manuever is reserved for stand-still and very slow speeds. Not having ridden it, I can only speculate.

I too have been curious about the turning behaviour and limitations at speed. Hopefully, some of those that have ridden it a bit more than you have will chime in with some viewpoints.


How much time did you have when you tried it out? Care to share your reflections on the experience with the board?

Charmed

Glider
09-13-2002, 03:08 PM
quote:Originally posted by charmed

Hiya Glider

you ask about motorcyclists' experiences with twist throttles. I have owned several over the years, and never had the problem you seem to fear. I did once have an ATV with a thumb lever throttle that took some getting used to (ie. hit a bump and vroom ahead) but this was easily mastered.

Regarding a sudden full twist of the steering control at speed--I always assumed that the circuitry would impose limits on turning relative to a given speed. In other words, moving slow, faster turning, moving fast, slower turning. Segway, from what I have seen, has arguably the world's most advanced interface for a ground based vehicle. I can't imagine that they would allow the rider to force the unit into a pirouette at 12.5 mph. when it would be so simple to program in proportional limits. More likely that manuever is reserved for stand-still and very slow speeds. Not having ridden it, I can only speculate.

I too have been curious about the turning behaviour and limitations at speed. Hopefully, some of those that have ridden it a bit more than you have will chime in with some viewpoints.


How much time did you have when you tried it out? Care to share your reflections on the experience with the board?

Charmed


I suspected the same control restrictions on violent turns, but then asked about the intentional violent turn and how would the Segway react.
My experience was too brief, although I often find myself reliving it as I walk the halls (7 to 8 minutes total before others got their chance). We would use it indoors for large scale integration projects that sprawl across large field sized buildings. There exists a lot of handicapped ramps that aren't used much but are the focal point for some of our business weenies that have boxes to check.

I did notice that the Segway only balances the rider in the forward and back directions, the rider needs to shift their wieght for lateral g forces.

I don't fear the effect, I just can't model it in my head. I can see getting startled and that involuntary movement back could induce unwanted control commands into the left right steering.

Seeker
09-13-2002, 04:03 PM
I wonder if there's any way for the Segway to be able to distinguish between 'genuine commands' and what you might call 'artifact commands',such as the input generated by a startle response.

It might be very challenging for a Segway to differentiate between valid and invalid responses, given the wide range of responses that are possible for different situations. But maybe if the Segway computers came across input data which had a high likelihood of being an artifact command, then they could 'smooth out' the response to this type of command (eg The Segway could respond, but respond more slowly than requested).

I'm just plain conjecturing here...I know nothing about any of this kind of stuff, based on know how, or experience. Could someone else tell us whether doing something like this might be possible ?

Seeker

Glider
09-13-2002, 04:11 PM
quote:Originally posted by Seeker



I wonder if there's any way for the Segway to be able to distinguish between 'genuine commands' and what you might call 'artifact commands',such as the input generated by a startle response.

It might be very challenging for a Segway to differentiate between valid and invalid responses, given the wide range of responses that are possible for different situations. But maybe if the Segway computers came across input data which had a high likelihood of being an artifact command, then they could 'smooth out' the response to this type of command (eg The Segway could respond, but respond more slowly than requested).

I'm just plain conjecturing here...I know nothing about any of this kind of stuff, based on know how, or experience. Could someone else tell us whether doing something like this might be possible ?

Seeker

Anyone know how complex the rider interface platform is telemetry wise? Is it just a weight sensor gate, or does it measure the riders Center of Mass?
If it did the latter, then the system would sense the rider leaning into the turn and conclude it wasn't an artifact, whereas a upright and otherwise relaxed rider would not want to fly head over the handlebars with such a command.

n/a
09-13-2002, 04:22 PM
quote:You do realize, Antagony (I LOVE that name) that Segway has undergone thousands of hours of testing in real world situations on the street and sidewalks of our great nation?

This is actually a very good point that has been overlooked. These tests are still going on. Hopefully the results will eventually be made available to the public. One weakness of these tests though is that postal workers etc. are likely to be a lot more carefull not to cause accidents while riding on a Segway than the average Joe since they could risk loosing their jobs if they ride wrecklessly. On the other hand the averge Joe might risk a lawsuit, not to mention personal injury.

That said I maintain my view that the more tests they do to prove Segway's safety the better. The lack of published safety tests has already cost them the use of sidewalks in Charleston. But we cant be sure that even if they had such data, they would have been banned from sidewalks because u cant guard against careless riders.

Seeker
09-13-2002, 06:45 PM
Hi Guys,

What would you consider to be the top 3 most dangerous situations, in terms of things a Segway rider could do, which would contribute towards an accident occurring ?

Seeker

Casey
09-13-2002, 06:59 PM
quote:The lack of published safety tests has already cost them the use of sidewalks in Charleston.

I have to disagree with that. In reading their ordinance it is apparent to me that they simply do not want "tourists" distracted or lulled into inattention while doing anything but walking. They have banned ALL wheeled vehicles from the sidewalks.

My guess is that eliminating ALL but medically approved devices from the sidewalks they are simply coveing their own butts against possible lawsuit.

Casey
09-13-2002, 07:08 PM
quote:Originally posted by Seeker



Hi Guys,

What would you consider to be the top 3 most dangerous situations, in terms of things a Segway rider could do, which would contribute towards an accident occurring ?

Seeker


1-Allowing one wheel to drop off a curb, or into a hole.
2-Riding longitudinally on sloped ground.
3-Riding down stairs. (I don't think you could ride UP)

Seeker
09-17-2002, 06:51 PM
Here's 2 of the dangerous situations we've identified, which Segways could be involved in :

-Riding longitudinally along sloped ground
-Looking to the side for long periods of time, as you drive

For the first situation, is there a way for the Segway to assess whether one of the wheels is too high, in relation to the other, and issue a warning signal about this ?

I recall reading something in the patents about the platform vibrating as a way of telling you that there was something amiss on the Segway.

As far as the danger associated with someone spending too much time looking to the side, here's a possible solution :

If the Segway could somehow determine from your 'lean patterns' that you were looking to the side, rather then looking forward, for a long period of time, it could issue you a warning (again the vibration of the platform might be the least intrusive way to warn you about this type of situation).


Can a Segway act in the same way that other measuring instruments such as lie detectors, or the equipment that measures earth quake activity, works ? ( Can it determine that a specific set of data ( eg lean patterns) represents a specific dangerous situation ? ( eg looking to the side for too long)

Seeker

baantjer
09-17-2002, 07:08 PM
quote:Originally posted by Seeker



Here's 2 of the dangerous situations we've identified, which Segways could be involved in :

-Riding longitudinally along sloped ground
-Looking to the side for long periods of time, as you drive

For the first situation, is there a way for the Segway to assess whether one of the wheels is too high, in relation to the other, and issue a warning signal about this ?
<snip>


Is appears to do well in this case: http://www.whatmagnet.com/Images/Content/segway200.gif

quote:Originally posted by Seeker


<snip>
I recall reading something in the patents about the platform vibrating as a way of telling you that there was something amiss on the Segway.
<snip>


If you lean too far forward something (the handle?) starts to vibrate to warn you that it has reached its topspeed.

Casey
09-17-2002, 07:49 PM
quote:Is appears to do well in this case:

I was actually picturing a sudden drop off the curb as opposed to riding on it. That does seem to simulate the longitudinal slope though.

I have always thought that the guy hurt in Atlanta probably hit a pothole or possibly a ridge of some kind.

Frank

Cube128
09-17-2002, 08:12 PM
Ahhh, I like this forum far better than TIQ already... Anyway, I think that most of this has probobly been thought up by Segway. I believe that the Segway's giroscoped can track any movement, either front and backward and side to side. I'm sure the Segway will warn yoy somehow if you are about to tip over. In regards to taking action to prevent that, I'm sure it's been left out for a reason.
Let's say you're riding on a crowded sidewalk in Sidewalk Mode, or 8 MPH. One of the wheels falls into the grates that cover the trees growing on the sidewalk, and you begin to tip. To correct this, the Segway would have to make a sharp turn left or right, possibly running into another pedestrian, creating a new problemb.
I believe that handling tipping issues will mostly be left to the rider, since most people have enough common sense to avoid them. Still, there will eventualy be the fool who takes a bad fall after tipping over, and sues Segway for 10s of millions of dollars.

ziggystardust_
09-17-2002, 08:39 PM
3 (or more)most dangerous things to do on a Segway:

1. Crank the yaw control hard left (or right) while moving forward at speed

2. Hitting a pothole and not adjusting your body to absorb the impact

3. Dropping a wheel off a curb at speed

4. Riding down stairs

5. riding headlong into oncoming automobile traffic

6. lose traction (i.e. on slippery ice or on gravel, etc.)


btw, when trying to exceed top speed, the control shaft will push back on you, causing you to lean back and thus slow you down so that you can't exceed maximum speed. The control shaft will shake when you either ride backwards too fast or the Segway identifies an error situation (such as the batteries running out of juice)

Seeker
09-17-2002, 09:30 PM
Here's another thing which I wonder whether the Segway may be capable of...

When you encounter a pedestrian on the sidewalk, I would think that your initial reaction would be to pull backwards (maybe just slightly), at least as you got closer anyways. Would there be a way for the Segway computers to 'know' that there must be a pedestrian in your path ( based on your pullback reaction) ? If the Segway determined that you had encountered a pedestrian, and yet you were still cruising along at a speed that was too fast, maybe it could vibrate, to warn you, until you slowed down to a speed that was reasonable for passing pedestrians.

If this were possible, I would think that it would reduce the occurence of accidents, and also reduce the severity of injuries resulting from them.

Seeker

ziggystardust_
09-17-2002, 10:12 PM
quote:Originally posted by Seeker


Here's another thing which I wonder whether the Segway may be capable of...

When you encounter a pedestrian on the sidewalk, I would think that your initial reaction would be to pull backwards (maybe just slightly), at least as you got closer anyways. Would there be a way for the Segway computers to 'know' that there must be a pedestrian in your path ( based on your pullback reaction) ? If the Segway determined that you had encountered a pedestrian, and yet you were still cruising along at a speed that was too fast, maybe it could vibrate, to warn you, until you slowed down to a speed that was reasonable for passing pedestrians.

If this were possible, I would think that it would reduce the occurence of accidents, and also reduce the severity of injuries resulting from them.

Seeker


programming software into the Segway that would try and 'guess' what your motivation is for performing an action would be very dangerous. There are other ways for a device to detect objects around it, and are in use right now, even in automobiles. Something like that would be vastly preferable to an HT guessing game

4Seg
09-17-2002, 10:52 PM
quote:Originally posted by Seeker



I wonder if there's any way for the Segway to be able to distinguish between 'genuine commands' and what you might call 'artifact commands',such as the input generated by a startle response.

It might be very challenging for a Segway to differentiate between valid and invalid responses, given the wide range of responses that are possible for different situations. But maybe if the Segway computers came across input data which had a high likelihood of being an artifact command, then they could 'smooth out' the response to this type of command (eg The Segway could respond, but respond more slowly than requested).

I'm just plain conjecturing here...I know nothing about any of this kind of stuff, based on know how, or experience. Could someone else tell us whether doing something like this might be possible ?

Seeker


Very valid questions!

No artifact movements. The Segway is as like Kamen says, like a new pair of sneakers. It's not like any experience you've ever had, trust me. One doen't "control" the segway per se' it glides with you. Your shoes don't run w/out you do they? Neither will the Segway. You can see where I tried to "control" IT at the very beginning... Click the middle movie clip (make sure you save it to disk first and that you have QuickTime 6)
www.stlsegway.com/movies


If we could only be resellers...

Seeker
09-18-2002, 12:18 AM
quote:

programming software into the Segway that would try and 'guess' what your motivation is for performing an action would be very dangerous. There are other ways for a device to detect objects around it, and are in use right now, even in automobiles. Something like that would be vastly preferable to an HT guessing game


Hi Zig,

I'm wondering why you would consider it to be dangerous, if the only response by the Segway to a situation identified as driving too fast under the circumstances, was to produce a caution vibration.

Seeker

4Seg
09-18-2002, 11:35 AM
quote:Originally posted by Seeker

quote:

programming software into the Segway that would try and 'guess' what your motivation is for performing an action would be very dangerous. There are other ways for a device to detect objects around it, and are in use right now, even in automobiles. Something like that would be vastly preferable to an HT guessing game


Hi Zig,

I'm wondering why you would consider it to be dangerous, if the only response by the Segway to a situation identified as driving too fast under the circumstances, was to produce a caution vibration.

Seeker

Seeker, The Segway actually produces a caution vibration when traveling backwards above 3mph. A Segway employee mentioned that the show. The vibration is simply "drive chatter" from the motors themselves I think. If I remember correctly I think it does the same thing forward. (leaning into it too much?) Brooster, do you remember anything like that?

If we could only be resellers...

See the Segway clips! http://www.stlsegway.com/movies

ziggystardust_
09-18-2002, 12:32 PM
quote:Originally posted by Seeker

quote:

programming software into the Segway that would try and 'guess' what your motivation is for performing an action would be very dangerous. There are other ways for a device to detect objects around it, and are in use right now, even in automobiles. Something like that would be vastly preferable to an HT guessing game


Hi Zig,

I'm wondering why you would consider it to be dangerous, if the only response by the Segway to a situation identified as driving too fast under the circumstances, was to produce a caution vibration.

Seeker


Sorry, I misread your statement. Yes, if the only thing the Segway would do is vibrate it's control shaft when it somehow 'sensed' you were coming too close too fast to a pedestrian, then that may be some type of solution. There are so many variables to that though. Would be difficult (and expensive) to program it for that.

ziggystardust_
09-18-2002, 12:34 PM
quote: If I remember correctly I think it does the same thing forward. (leaning into it too much?) Brooster, do you remember anything like that?



Actually, the Segway does not vibrate when you lean too far forward. Leaning forward causes the wheel to turn faster (via the motors) to 'catch up' to you. When you hit a certain speed (the top speed as dictated by the smart key), the control shaft will 'push back' on you, causing you to lean back and thereby slow you down so that you don't exceed the set limit.

Seeker
09-18-2002, 01:04 PM
quote: There are other ways for a device to detect objects around it, and are in use right now, even in automobiles.


What other types of detection systems do you think might be good ones for Segway to look into at some point ?

Seeker

ziggystardust_
09-18-2002, 04:36 PM
quote:Originally posted by Seeker

quote: There are other ways for a device to detect objects around it, and are in use right now, even in automobiles.


What other types of detection systems do you think might be good ones for Segway to look into at some point ?

Seeker

I'm not an engineer so I don't have a good grasp on the subject, but I do know that some cars have sensing devices that will warn if you're about to back up into another car. Maybe someone wants to do some research on this stuff