1.  Difficult to implement, but a good idea.

2.  Car headlights are far from sufficient - illuminating something at any distance requires a beam angle that coincidentally blinds oncoming traffic.  Furthermore, a motion sensor approach that turns off, say, any half-mile stretch of road without cars on it, is unfortunately impossible.  The high intensity discharge lamps (10x the efficiency of halogens, and easily focussable using reflectors) used in streetlamps are notoriously averse to on/off cycles - it takes a good half hour for the low pressure sodium lamps at the parking lot I use to warm up to usable temperature, and another half hour to cool down after shutting off to the point that they can be restarted without bursting from the heat load(sounds contradictory, but extremely high temperatures are involved).

3.  These should be implemented, and in the near-desert majority of the American West, could (politically) be implemented this moment if you pitch the right people.

4.  Drip irrigation is one of the essential techniques in green/lowenergy agriculture.  Subsurface water carriers are another.  Putting the water directly into the ground without any opportunity for evaporation, six inches from the plant it's targetting, is loads more efficient than spraying it into the air.  Underground, it can then catch on a buried drainpipe section and remain there saturating the soil until the plant's rootstructure reaches it.

5.  Auto-brake(collision detection) can and is used now, but it' a race between dangerously preempting the driver for longer range threats that the driver is steering around, and not having the stopping distance to avoid a collision.  Auto-Swerving is of questionable legality, as it can easily cause an accident in and of itself.

6.  Fantasy of biology.  Never gonna happen.
   
7.  While tasers should be considered semilethal force, a much lower voltage stun gun placed in an ankle bracelet could be implemented as you say to great success.   A prisoner who can't stand up because of calf convulsions is a prisoner who can't continue a fight, without needing to have a guard confrontation or firearms involved.

I think a much more important development is decriminalizing nonviolent drug offenses.

8.  Fantasy of human nature.  Never gonna happen.  In other news, emergency sirens have showed particularly good efficiency when playing tracks involving chopped up human screams.

9.  This is how the price of depletable resources is supposed to work in a sufficiently large free market.  Utilities, however, are neither.  Good luck getting it accepted in the political arena.

10. Not gonna happen, and besides, not needed:  the overweight and obese already pay more for personal transportation, insurance, etc.

#3 - Mandatory urinals?  So, if it's purchased by a woman who lives alone or only with other women, she has to keep this piece of useless plumbing?  (Male guests?  Big deal - you're STILL requiring a square meter of floorspace for insignificant use.)  I'll even add to this wtf remark - go into the bathroom in most houses and try to figure out where you're going to put a urinal.  Unless you're buying a somewhat upper scale house, bathroom space already has dense utilization.

#7 (both totoneila and squalish) I highly recommend you spend some time working in a prison before you make blithe assumptions as you've just done.  Let's start with the practical:
a) How do you intend to keep this bracelet charged?  Remember it's got to be sensing other bracelets 100% of the time.
b) Sensing - how do you avoid false discharges?  Given that some of the inmates will be working around heavy machinery, there are times when falling with an uncontrollable calf spasm would be a very bad idea.
c) So you've solved the power and sensing problem, now how do you keep this bracelet on the inmate?  (If the answer in any way involved surgery, stop and re-evaluate.)  This one is REAL important.  Inmates make weapons out of things you would not believe.  Please remember that for the inmate time is not an issue.  So what you're doing is giving the inmate a single-shot stunner.  If he can get it off he can use it on someone else - like a correctional officer/guard.
d) Note in c I said single-shot.  Actually, one stun charge is probably not going to be enough.  First shot causes a cramp.  Unless response is immediate, the inmate can compensate and recover and continue with the action.  So what you actually need is multi-shot, except now you've put a multi-shot weapon in the hands of the inmate facing the officers.
e) off the stunner, on to a comment from Squalish - Yes, with a lot more surveillance and several other techniques many of the problems could be severely reduced.  Not eliminated - sorry - but certainly severely curtailed.  The big problem with this is money.  The second largest problem is politics.  sigh SOME prisons rely on inmate violence as part of their control.  MOST prisons (at least in the US) do not.  Trust me on this, I spent long enough working in that field to know this.  The primary problem with inmate violence is that violence unleashed is difficult to control - which in turn means it splashes on the prison staff.  There are other very expensive problems with using misbehavior as a behavioral control as well - expensive in social and monetary terms.  There are a lot of problems with the so-called correctional system, but most of them are the result of having to deal with people.  (Mencken is RIGHT.  Any obvious and simple solution to a human problem is almost certainly wrong.)