Velocity PID is the algorithm typically used to perform closed-loop velocity control on a position or velocity axis. PID stands for the central gains used in this mode: Proportional, Integral, and Differential. The Velocity PID provides very good control and is suitable for nearly all motion control systems with velocity feedback. In certain cases, Velocity I-PD control may be preferred.
Velocity PID Advantages
Excellent for controlling an axis that follows a smooth target, such as one generated by the RMC motion commands.
Velocity PID Disadvantages
May not control very well with an irregular target, such as step jumps or a joystick.
Motion Commands in Velocity PID Mode
To use Velocity PID, use the Set Pos/Vel Ctrl Mode (68) command to set the Next Pos/Vel Control Mode to Vel PID. The next closed-loop motion command will use the control mode specified in the Next Pos/Vel Control Mode status register. The Current Control Mode register indicates the mode currently in use.
See the Closed Loop Control topic for details on which commands are supported in Velocity PID control.
Each closed loop motion command issued to the RMC specifies a target profile, which defines where the axis should be at any given moment. For each loop time when the axis is in closed loop control, the RMC uses the specified target profile to calculate the desired velocity of the axis at that moment (called the Target Velocity) and subtracts the Actual Velocity to determine the Velocity Error. The Velocity PID algorithm then uses this information, together with the gains and feed forwards, to calculate how much Control Output should be generated to move the axis to the Target Velocity. The values of the gains and feed forwards must be set to achieve proper control. The process of setting the gains is called tuning and is done as part of the setup procedure.
The Velocity PID uses the gains and feed forwards listed below. Each gain or feed forward is multiplied by some quantity related to the Target Velocity and Actual Velocity to come up with a percentage. The resulting percentages are all summed and then multiplied by the maximum output (typically 10V), to come up with the Control Output voltage for that loop time.
The Proportional Gain is multiplied by the Velocity Error. This is the most important gain.
The Integral Gain is multiplied by the integrated (sum of value x time) Velocity Error. This helps the axis get to velocity over time.
The Differential Gain is multiplied by the difference between the Target and Actual Accelerations. This helps the axis keep up with quick changes in velocity.
Velocity Feed Forward
The Velocity Feed Forward is multiplied by the Target Velocity.
Acceleration Feed Forward
The Acceleration Feed Forward is multiplied by the Target Acceleration.
Jerk Feed Forward
The Jerk Feed Forward is multiplied by the Target Jerk. The Jerk Feed Forward is not necessary for most applications.
Tuning Velocity PID
The velocity PID gains must be tuned manually. The Tuning Wizard cannot be used to tune velocity PID control.
Copyright (c) 2005-2012 by Delta Computer Systems, Inc.