Nonholonomic Path Planning
		--------------------------
Shyjan Mahamud		  Wed 17th Jan 1996		Wean 8220

I describe a method for planning obstacle-avoiding paths for robots which have
nonholonomic constraints. In the simplest case, these constraints model the 
fact that in reality the velocity of the robot is restricted to a linear
subspace of their configuration space. The method is based on the use of
harmonic functions. Linear nonholonomic constraints can be directly expressed 
as boundary conditions for the harmonic function. Harmonic functions can be
readily implemented in a resistive network. I describe both the theory and it's
implementation.