An electronic controller is considered the brain of a robot. It is actually similar to a computer, which stores data relating to the robot and work environment as well as executing programs in order to operate the robot. The control system includes programs, data algorithms, logic analysis, programs and a number of other activities helping the robot to perform.
Industrial controllers are either non-servos, point-to-point servos or continuous point servos. A pick and place robot or non-servo robot moves components from one place to another. The movement of the non-servo robot is manipulated by the controller. The controller receives a signal from the stop switch and the next movement is started.
Advanced control systems are present in advanced robots. Two electronic boards connected with flex cables comprise the Mars Soujourner rover brain: one is the power board, the other is the CPU board comprising items for power generation, power conditioning, analog and digital I/O control and processing, power distribution, and data storage.
It is possible to operate mobile robots by remote control or autonomously. Instructions are provided by a human operator to a remote control robot. Pre-programming of autonomous robots is fundamental to understanding their environment and to receive a command independently based on the knowledge they possess. Some autonomous controllers learn from previous encounters. They can identify a situation, process actions that have delivered unsuccessful or successful results and modify their behavior for success optimization. These activities are performed in the controller.
A galvanic skin response sensor was used by Meisner E et al from the Department of Computer Science, Rensselaer Polytechnic Institute in 2007 as a tool to develop controllers for interaction between humans and robots. An algorithm was presented by the team that yields human-friendly controllers.
In future, this team will study using biofeedback in a controller design for conducting larger, more controlled experiments involving complex tasks. The use of further biofeedback sensors such as heart-rate sensors will be explored by the team for the design of better controllers.
Peri V.M et al conducted a research on the controller of an autonomous wall robot. Mounted ultrasonic sensors provide the input for the robots. A Microchip PIC16F877 controller on board the robot studies the data and emits the needed control signal. The robot’s motion is controlled using a fuzzy logic controller along a predefined path. Initially, the robot was modeled in Matlab Simulink and the fuzzy logic rules optimized for best results. The microcontroller was later programmed in C language with a PCW-C compiler and tested. The controller performance was evaluated in the study. It will be possible to use this kind of controller with minor modifications in a number of real world applications such as building exploration in disaster areas, autonomous cars and inter-office mail delivery. The key improvement of the system will be to use a faster, better microcontroller so that fuzzy calculations can be done on board instead of a lookup table.
A good example of advanced controllers for robotics comes from the Institute for Computer Science Technische Universitat Bergakademie Freiberg. This institute has developed an advanced kinetic controller for humanoid robots.
Advanced Kinect Controller for Humanoid Robots
Controllers are required in every robot and are presently used in a wide range of robotic applications. With further advancements in the design and development of controllers, it will be possible to venture into novel areas within the field of robotics and automation.
Manufacturers are starting to integrate robotic technology into a work cell. The advancement appears to be influenced by the ideal of easy use of the robot software, improved capability of the robot to perform functions normally done by external devices.
The idea to downsize controllers will continue for better application of robots in life science installations and a laboratory environment. Being able to miniaturized controllers allows for a more compact robotic structure that can be safer for non-industrial applications.
The design and development of advanced controllers allows for a much wider scope of application areas for robotic technology especially for the manufacturing industry when considering precision assembly and surface finishing.
Incorporating sensory inputs such as scanning lasers, sonar and 3D vision systems into a controller will start to expand the target industries for this technology. With improved safety circuitry and smarter controllers, the future of robots has the opportunity to face challenges such as the maintenance of ship components and work on oil rigs.
Sources and Further Reading
- The Rover Raunch Presents Robot Systems - NASA
- Meisner E, et al. Controller Design for Human-Robot Interaction. 2007. CiteSeer.