In my $Theoretical \ Mechanics$ book I recently read, it gives the following description about "the path" in a certain time interval:

$$\vec{r}(t)=\vec{r}(t_0)+\vec{v}(t_0)(t-t_0)+\int_{t_0}^tdt^{\prime}\int_{t_0}^{t^{\prime}}dt^{\prime \prime}a(t^{\prime \prime})$$

where the quantities $r(t_0)$ and $v(t_0)$ are the positions and speed at time $t_0$, the initial conditions.

While $\vec{r}(t_0)+\vec{v}(t_0)(t-t_0)$ is fairly obvious, I am having in difficulty understanding why it puts so many derivatives and integrations in $\int_{t_0}^tdt^{\prime}\int_{t_0}^{t^{\prime}}dt^{\prime \prime}a(t^{\prime \prime})$. Can anyone help me to explain what does it mean? (I know it stands for the integration for the acceleration component, but why it represents it like this?)

  • 2
    $\begingroup$ so many derivatives and integrations The primes do not indicate derivatives. $t’$ and $t’’$ are just other time variables. There are only two integrations happening here. $\endgroup$
    – G. Smith
    Commented Aug 3, 2019 at 16:46

1 Answer 1


As G. Smith suggests in the comments, the primes don't indicate derivatives, they're just used to distinguish the new integration variables $t',t''$ with $t$. To see where this comes from, look at the definition of acceleration:

$$\frac{d^2}{dt^2} \mathbf r(t) \equiv \mathbf a(t)$$

Now integrate both sides with respect to time from $t_0$ to $t$:

$$\int_{t_0}^t \frac{d^2}{dt_1^2} \mathbf r(t_1) dt_1 = \int_{t_0}^t \mathbf a(t_1) dt_1$$

where I've used the integration variable $t_1$ instead of $t'$ to avoid confusion with derivatives. Using the fundamental theorem of calculus, this gives:

$$\mathbf v(t) - \mathbf v(t_0) = \int_{t_0}^t \mathbf a(t_1) dt_1$$

with $\mathbf v \equiv d \mathbf r/dt$ being by definition the velocity. Integrating both sides again similarly results in:

$$\int_{t_0}^t \mathbf v(t_1) dt_1 - \int_{t_0}^t\mathbf v(t_0) dt_1 = \int _{t_0}^t \int_{t_0}^{t_1} \mathbf a(t_2) dt_2 dt_1$$

$$\int_{t_0}^t \frac{d\mathbf r(t_1)}{dt_1} dt_1 - \mathbf v(t_0) \int_{t_0}^t dt_1 = \int _{t_0}^t \int_{t_0}^{t_1} \mathbf a(t_2) dt_2 dt_1$$

$$\mathbf r(t) - \mathbf r(t_0) - \mathbf v(t_0)(t-t_0) = \int _{t_0}^t \int_{t_0}^{t_1} \mathbf a(t_2) dt_2 dt_1$$

Rearranging this gives the final result:

$$\mathbf r(t) = \mathbf r(t_0) + \mathbf v(t_0)(t-t_0) + \int _{t_0}^t \int_{t_0}^{t_1} \mathbf a(t_2) dt_2 dt_1$$

As Chet Miller points out in the comments, the double integral can be further simplified by using integration by parts:

$$\int _{t_0}^t \int_{t_0}^{t_1} \mathbf a(t_2) dt_2 dt_1 = \int _{t_0}^t \left(\int_{t_0}^{t_1} \mathbf a(t_2) dt_2\right) dt_1 = t_1 \left(\int_{t_0}^{t_1} \mathbf a(t_2) dt_2\right) \Bigg\vert_{t_1=t_0}^t - \int_{t_0}^t t_1 \ \mathbf a(t_1) dt_1$$

$$= \int_{t_0}^t (t-t_1) \ \mathbf a(t_1) dt_1$$

This gives the simplified result:

$$\mathbf r(t) = \mathbf r(t_0) + \mathbf v(t_0)(t-t_0) + \int_{t_0}^t (t-t_1) \ \mathbf a(t_1) dt_1$$

  • 1
    $\begingroup$ Shouldn't the final answer involve $\mathbf{a}(t_2)$, not $$\mathbf{a}(t_1}? Otherwise, it is not consistent with the original post. Also, it seems that this can be reduced to a single integral by integrating by parts. $\endgroup$ Commented Aug 4, 2019 at 0:36
  • $\begingroup$ @ChetMiller That's right! Thanks for correcting me. True, the final double integral can be simplified. I just didn't bother since that wasn't part of OP's question. $\endgroup$ Commented Aug 4, 2019 at 0:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.