Wrong Model of projectile motion ?

[nicolapace]

The model for 1D-projectile motion is the following:

x_(k+1) = x_k + v_k * dt + 0.5 * a_k * dt^2
v_(k+1) = v_k + a_k * dt
a_(k+1) = a_k

that in matrix from is the following:

1		dt		0.5*dt^2
0		1		dt
0		0		1

That is different from the matrix implemented in the example

  // Discrete LTI projectile motion, measuring position only
  A << 1, dt, 0, 0, 1, dt, 0, 0, 1;

[bsyy6]

I agree with you. they really need to address that.

[nicolapace]

any news @hmartiro ?

[ghornig]

We assume the time difference is small (i.e. the limit dt->0.) In which case dt^2 << dt and that term can be neglected.