Robotis Dynamixel X Series, Руководство пользователя

Страница: 17 / 34

NOTE : K
a
stands for Anti-windup Gain and β is a conversion coefficient of position and velocity that
cannot be modified by users. For more details about the PID controller, please refer to the PID
Controller at wikipedia .
2. 4. 24. Position PID Gain(80, 82, 84), Feedforward 1st/2nd Gains(88, 90)
These Gains are used in Position Control Mode and Extended Position Control Mode. Gains of
DYNAMIXEL’s internal controller can be calculated from Gains of the Control Table as shown
below. The constant in each equations include sampling time. Position P Gain of DYNAMIXEL’s
internal controller is abbreviated to K
P
P and that of the Control Table is abbreviated to
K
P
P
(TBL)
.
Controller
Gain
Conversion
Equations
Range Description
Position D Gain(80)
K
P
D K
P
D = K
P
D
(TBL)
/ 16
0 ~
16,383
D Gain
Position I Gain(82)
K
P
I
K
P
I = K
P
I
(TBL)
/
65,536
0 ~
16,383
I Gain
Position P Gain(84)
K
P
P K
P
P = K
P
P
(TBL)
/ 128
0 ~
16,383
P Gain
Feedforward 2nd
Gain(88)
K
FF2nd
K
FF2nd(TBL)
/ 4
0 ~
16,383
Feedforward Acceleration
Gain
Feedforward 1st
Gain(90)
K
FF1st
K
FF1st(TBL)
/ 4
0 ~
16,383
Feedforward Velocity Gain
Below figure is a block diagram describing the position controller in Position Control Mode and
Extended Position Control Mode. When the instruction from the user is received by
DYNAMIXEL, it takes following steps until driving the horn.
1. An Instruction from the user is transmitted via DYNAMIXEL bus, then registered to Goal
Position(116) .
2. Goal Position(116) is converted to desired position trajectory and desired velocity trajectory
by Profile Velocity(112) and Profile Acceleration(108) .
3. The desired position trajectory and desired velocity trajectory is stored at Position
Trajectory(140) and Velocity Trajectory(136) respectively.