Approach to Decomposing Joint Transformation Matrix Extracting Joint to Joint transformation matrices from the robot joint vectors which include a position vector and a compound transformation matrix. YA A θ3 {C} {D} P=(x,y,z) θ4 1 θ2 {B} A θ1 PDORG XA {A} Step 1: Set a point vector #P to the home position via #P=#PPOINT(0,0, …,0). Step 2: Execute the ROTJOINT V+ program which outputs a compound transformation matrix for the point #P. Set the rotation angle to 30° just for the joint in question, with all others to zero. Step 3: Set up transformation equations and successively evaluate the LHS with the RHS from the program output and the prior calculations. T ABT 1 CAT B C T CAT 1 DAT CAT 1 ABT CBT C D Step 4: Replace the non-zero or one numeric values in the rotation matrix portion of the transformation with either sinθ = 0.50 or cosθ=0.866. Alternative: Manually generate all joint frame work in trigonometric equation form and compare the results by entering θ = 0°, 30°, 60°, 90°, etc. for selected joints. Adept 550 SCARA robot joint configuration The Adept 550 follows the convention in frame structuring, namely, the joint rotation or translation is about or along Z axis and joint-to-joint frame rotation is done about X axis. The following is a complete set of transformation matrices which has been numerically verified with ROTJOINT. Joint 1 z Joint 2 z c1 R R R R R s1 0 0 1 1 2 2 3 3 4 c12 R s12 0 s12 c12 0 x s1 0 c2 c1 0 s2 0 1 0 c12c4 s12s4 c12s4 s12c4 0 0 4 Joint 4 x y x 0 4 Joint 3 s2 c2 0 y x z z 0 1 0 0 c4 0 0 1 0 s4 1 0 0 1 0 s4 c12s4 s12c4 c12c4 s12s4 0 c4 0 0 c124 2s12s4 0 s124 2c12s4 1 0 0 s12 0 0 when θ4=0. 4 R c12 0 1 c12 s12 0 0 0 1 s124 2c12s4 c12c4 2s12s4 0 0 0 1 0 0 when θ4=90°. 1 The Adept transformation vectors carry the compound values of 04T . The 4th column vector is 0 P4org L1c1 L2c12 L1s1 L2 s12 D d3 T where L1=300 mm, L2=250 mm, and D is the maximum reach in Z direction. The model number “550” comes from L1+L2. Adept Six 300 robot joint configuration Assuming that the Adept six axis robot follows the convention in frame structuring, namely, the joint rotation or translation is about or along Z axis and joint-to-joint frame rotation is done about X axis. The following is a set of the transformation matrices which has been partially verified with ROTJOINT. Joint 1 Joint 2 Joint 3 z x x x y s1 0 c1 0, 0 1 1 0 0 c2 s2 1 0 1 s2 c2 2 R 0 0 1 0 0 0 c3 s3 0 2 c3 0 3 R s3 0 0 1 1 0 0 c4 s4 0 1 s 3 4 R 0 4 c4 0 1 0 0 0 1 0 0 c5 s5 4 0 1 s5 c5 5 R 0 0 1 0 0 0 1 0 0 c6 s6 0 1 s 5 6 R 0 6 c6 0 1 0 0 0 0 1 Joint 5 Joint 6 z z y y x y y x x c1 R s1 0 0 c2 0 0 1 s2 0 c4 0 0 1 s4 0 c5 0 0 1 s5 0 c6 0 0 1 s6 c1c23 c1s23 s1 0 0 1 2 s1s23 c1 3 R 1 R 2 R 3 R s1c23 s23 c23 0 0 6 Joint 4 z z y z R 30R 63R s2 0 c2 s4 0 c4 0 1, 0 0 1, 0 s5 0 0 1, c5 0 s6 0 0 1, c6 0 c4c5c6 s 4 s6 3 3 4 5 s5c6 6 R4 R5 R6 R s4c5c6 c 4 s6 c4c5 s6 s 4 c6 s5 s6 s4c5 s6 c 4 c6 c4 s5 c5 s4 s5