Question 4 of 7: Frame assignment, DH parameters and forward kinematics of the RPR arm (20 marks)
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exams, December 2018 — 16-Mec-B12 Robotics. Closed book, three hours, one 8.5″×11″ two-sided formula sheet and an approved Casio or Sharp calculator permitted. Seven questions, each of equal value (20 marks); the rubric states that five questions constitute a complete paper and that only the first five in the answer book are marked. As a study resource all seven are solved here. Marks shown in parentheses are the printed sub-part values.
Notation. The paper's own nomenclature is used throughout: $\hat{X}_A,\hat{Y}_A,\hat{Z}_A$ are the unit vectors of frame $\{A\}$; a leading superscript as in ${}^{A}V$ names the frame the vector is referenced to; and ${}^{B}_{A}T$ is the transform of frame $\{A\}$ expressed relative to $\{B\}$. This is Craig's convention, so Craig's modified (distal) Denavit–Hartenberg parameters are used for the link tables.
Reference texts.
Craig, J. J., Introduction to Robotics: Mechanics and Control, 4th ed., Pearson, 2018 — the notation, the Z–Y–Z Euler convention and the two-segment cubic spline of Question 7 all follow this text (Ch. 2, 3, 5 and 7).
Spong, M. W., Hutchinson, S. and Vidyasagar, M., Robot Modeling and Control, 2nd ed., Wiley, 2020 — alternative treatment of DH assignment and Jacobian singularities (Ch. 3, 4).
Niku, S. B., Introduction to Robotics: Analysis, Control, Applications, 3rd ed., Wiley, 2020 — worked frame-graph and trajectory examples.
Siciliano, B., Sciavicco, L., Villani, L. and Oriolo, G., Robotics: Modelling, Planning and Control, Springer, 2009 — differential kinematics and statics duality (Ch. 3).
Question 4: Frame assignment, DH parameters and forward kinematics of the RPR arm (20 marks)
Given. The R–P–R arm of Figure 1: a revolute joint $\theta_1$ about the vertical base axis, a prismatic joint $d_2$ sliding parallel to that axis at a horizontal offset $l_1$, and a revolute joint $\theta_3$ whose axis is horizontal and perpendicular to the offset link. The figure states that all link lengths beyond joint 3 are zero, so the wrist centre $e$ coincides with the joint-3 origin.
Find. (a) a frame assignment, (b) the DH parameter table, and (c) the forward-kinematic transform ${}^{B}_{W}T$ from the base frame to the wrist frame.
[Figure not reproduced: Figure 4.1 — the arm as drawn on the paper, redrawn in elevation. The base revolute joint turns about the vertical, the slider carries the wrist up and down, and joint 3 turns about a horizontal axis. See the official exam paper.]
Approach. Attach a $z$ axis to every joint axis, place the $x$ axes along the common normals, read the four Craig (modified) DH parameters off the sketch, and multiply the three link transforms — then add the constant roll that reconciles the DH wrist frame with the triad drawn on the figure.
Affix the frames (part a). Following Craig's rules: $\hat{Z}_i$ lies on joint axis $i$; $\hat{X}_{i-1}$ lies along the common normal from $\hat{Z}_{i-1}$ to $\hat{Z}_i$; and the base frame $\{B\} = \{0\}$ is chosen coincident with $\{1\}$ when $\theta_1 = 0$. That gives $\hat{Z}_1$ vertical at the base, $\hat{Z}_2$ vertical at the far end of the offset link (the slider axis is parallel to the base axis), and $\hat{Z}_3$ horizontal along the joint-3 axis. Because $a_2 = 0$ and $d_3 = 0$, frames $\{2\}$ and $\{3\}$ share the wrist origin.
Figure 4.2 — part (a): frames affixed. $\{0\}$ and $\{1\}$ sit at the base; $\{2\}$, $\{3\}$ and the wrist frame $\{W\}$ share the wrist centre $e$. The joint-3 axis (marked with a cross) runs into the page, parallel to $\hat{Y}_0$ at $\theta_1 = 0$.
Read the DH parameters (part b). With frames placed, each line of the table is read directly off the sketch: $a_{i-1}$ is the distance along $\hat{X}_{i-1}$ between the two $z$ axes, $\alpha_{i-1}$ the twist about $\hat{X}_{i-1}$, $d_i$ the offset along $\hat{Z}_i$ and $\theta_i$ the joint angle about $\hat{Z}_i$.
Modified (Craig) DH parameters for the RPR arm; the joint variable of each row is shown in bold
Reconcile frame $\{3\}$ with the drawn wrist triad. The figure labels the wrist axes with $z$ pointing up and $x$ along the arm, whereas the DH rules force $\hat{Z}_3$ onto the joint axis (horizontal). The two differ by a constant roll of $+90^\circ$ about $\hat{X}_3$, which is a fixed property of the tool mounting and not a joint variable:
$$ {}^{3}_{W}T = \begin{bmatrix} 1 & 0 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 1 & 0 & 0 \\ 0&0&0&1 \end{bmatrix} = R_X(90^\circ). $$
No $l_3$ appears, exactly as the question notes.
Multiply out the kinematic equations (part c). Chaining the four factors,
$$ {}^{B}_{W}T = {}^{0}_{1}T\ {}^{1}_{2}T\ {}^{2}_{3}T\ {}^{3}_{W}T = \boxed{\begin{bmatrix} c_1c_3 & -s_1 & c_1s_3 & l_1c_1 \\ s_1c_3 & c_1 & s_1s_3 & l_1s_1 \\ -s_3 & 0 & c_3 & d_2 \\ 0 & 0 & 0 & 1 \end{bmatrix}}. $$
The wrist centre is therefore $[x_e,\ y_e,\ z_e]^{T} = [\,l_1c_1,\ l_1s_1,\ d_2\,]^{T}$.
Sanity-check the structure. Three independent checks all pass. (i) The rotation block factors as $R_Z(\theta_1)R_Y(\theta_3)$, which is what the mechanism does — turn about the vertical, then about a horizontal axis. (ii) The wrist centre always lies at radius $l_1$ from the base axis and at height $d_2$, so the reachable surface is a cylinder of radius $l_1$: the arm cannot change its reach, only its azimuth and height. (iii) At $\theta_1 = \theta_3 = 0$ the transform reduces to a pure translation, i.e. $\{W\}$ is parallel to $\{B\}$, matching the figure.
Numerical spot-check. For $l_1 = 0.500$ m, $\theta_1 = 30^\circ$, $d_2 = 0.750$ m and $\theta_3 = -40^\circ$,
$$ {}^{B}_{W}T = \begin{bmatrix} 0.6634 & -0.5000 & -0.5567 & 0.4330 \\ 0.3830 & 0.8660 & -0.3214 & 0.2500 \\ 0.6428 & 0 & 0.7660 & 0.7500 \\ 0&0&0&1 \end{bmatrix}, $$
whose position column $[0.4330,\ 0.2500,\ 0.7500]^{T}$ m is exactly $[l_1\cos 30^\circ,\ l_1\sin 30^\circ,\ d_2]^{T}$.
Check — engineering assumptions. Two readings of Figure 1 are fixed here and should be stated on the answer paper as the rubric invites. First, the joint-3 axis is taken to be horizontal and perpendicular to the offset link (it is drawn parallel to $\hat{Y}_0$); this is what makes the arm an R–P–R with a pitching wrist. Second, the prismatic displacement $d_2$ is measured from the plane of the base frame, and all link dimensions between joint 3 and the wrist centre are zero as the figure annotation states. If a marker instead measures $d_2$ from the top of the slider housing, only the constant offset in the $(3,4)$ entry changes.
Question 4 — results
Part
Item
Result
(a)
Frames
$\hat{Z}_1$ vertical at base; $\hat{Z}_2$ vertical at offset $l_1$; $\hat{Z}_3$ horizontal; $\{2\}=\{3\}$ at the wrist