A control system is a mechanism that can be programmed to control robotic movements. It features logic analysis, data algorithms, programs and other processing methods that allow robots to perform tasks.
The control of robots involves three stages: perception, processing, and reaction. The control unit collects information provided by the sensors about the robot or environment, processes it, and and translates it into the required actuator commands.
The following are the basic methods of controlling robots:
Adaptive control method – This method manages the process model with respect to the prior actions, using feedback signals. Updating the results of prior actions helps in correcting the errors in the model due to changes in the environment for a long time period. However, this model is not suitable for correcting errors due to local changes in the environment.
Feed-forward control method – This control evaluates the amount of energy or extent of action required for performing a task by a robot. It is generally used for fixing the actuator settings in processes with less local disturbances and processes with delayed feedback signals.
Feedback control method – Feedback control checks the speed and position of the sensors by adjusting the power transmitted to the sensor. This method also fixes errors in positioning.
- Open loop control method – This is a basic control method that can be applied to systems carrying simple loads. In this method, the changes in a physical parameter such as torque, velocity, force or position required to perform the task is calculated. Following this, the amount of energy required by the actuator to carry out the change is determined.
Sources and Further Reading
- Basic Concepts of Robot control – Robot Basics