Consider an LTVC System: {x˙(t)=A(t)x(t)+B(t)u(t)y(t)=C(t)x(t)+D(t)u(t)x(t)∈Rn,u(t)∈Rmwhere A(t)∈Rn×n,B(t)∈Rn×m,C(t)∈Rp×n,D(t)∈Rp×m.
Let’s assume we know A,B,C,D and u(t)∣[t0,t1],y(t)∣[t0,t1]. Now we recall by Linear Time Varying Control System that x(t)=ΦA(t,t0)x0+t0∫tΦA(t,τ)B(τ)u(τ)dτ By our assumption, if we know the initial state, we can find x(t). Hence x0 is the unknown in the observability problem, and it is what we want to recover. We can express y as follows: y(t)y(t)y(t)=C(t)x(t)+D(t)u(t)=C(t)ΦA(t,t0)x0+t0∫tΦA(t,τ)B(τ)u(τ)dτ+D(t)u(t)=unknownC(t)ΦA(t,t0)x0+knownt0∫t1C(t)ΦA(t,τ)B(τ)u(τ)dτ+D(t)u(t)hence we can rearrange and write C(t)ΦA(t,t0)x0=y(t)−t0∫t1C(t)ΦA(t,τ)B(τ)u(τ)dτ+D(t)u(t)Then we define the linear mapL^:Rn→C([t0,t1;Rp])x↦H(t)x=C(t)ΦA(t,t0)x