A particle follows a discrete-time trajectory in R2 given by
(xt+1yt+1)=(1011)(xtyt)+(t1)ut+(ϵt0)
where {ϵt} is a white noise sequence with Eϵt=0 and Eϵt2=v. Given (x0,y0), we wish to choose {ut}t=09 to minimize C=E[x102+∑t=09ut2].
Show that for some {at} this problem can be reduced to one of controlling a scalar state ξt=xt+atyt.
Find, in terms of x0,y0, the optimal u0. What is the change in minimum C achievable when the system starts in (x0,y0) as compared to when it starts in (0,0) ?
Consider now a trajectory starting at (x−1,y−1)=(11,−1). What value of u−1 is optimal if we wish to minimize 5u−12+C ?