Walking with a constant velocity would be very difficult over a period of even a few seconds. You have to take into account that a truely constant velocity would require 100% accuracy. That is, with no rounding of time. There's a difference between walking 1 m/s and walking 1.000324 m/s, and this isn't true constant velocity. I don't think humans have the accuracy needed to walk at a constant velocity. z-component
Like wise constant acceleration would be impossible to maintain for any but very short periods of time. If you consider your velocity over significantly large distances you will probably find that constant velocity is a more reasonable answer. On a small scale, like a singe stride of course your velocity is constantly changing, but so is your acceleration, so you cannot call it constant anything on that scale. If you look at anything over 2 or 3 strides a constant velocity is much more likely then a constant acceleration. Even when you are running you only accelerate for the first few strides. I am sure someone will claim me incorrect because of some Olympic sprinter, just broaden your definition of "a few"!
well it would be impossible to accelarate constantly since we would have limits on how fast we can go, so it would sorta look like a square root graph, and we could only maintain a high speed for so long before we would get tired. however if you look at it this way, you measure the distance traveled, say 1 Km, and it took you one hour to get there, you would then be traveling at one km per hour, then you would pick random points, many of them, and measure your velocity at that point, and if it's 1km/hour most of the time then you are maintainting a constant velocity. on a diffrent note, say you are counting half notes, and every half note you put one foot infront of the other, this would come very close to getting you a constant velocity, either way its' more pheasable to keep going the same speed than to keep accelarating at the same rate.