The Shortcomings of Computer-controlled Robots

Employing computers in automation gave rise to what we now acknowledge as computer-controlled robots. They are machines integrated into a self-governing system that is controlled by interfacing them with personal computers—the integral parts of the robots that contain both the control and the task programs that the robots need to function.

Unlike the task programs of Artificial Intelligent (AI) robots that are contained in the robots themselves, the task programs of computer-controlled robots are provided by human operators who instruct the robots regularly on how to perform a particular task.

Regular Programming

As a consequence, regular programming is indispensable to controlling every work of computer-controlled robots. It is through it that an operator specifies what task such robots should do and how the various components must function to accomplish the task. To accomplish this, an operator needs to write the task programs. Since a task program can only consist of a limited number of well-defined actions that the robots must perform continuously and repeatedly in a particular sequence without deviation from one cycle to the next, computer-controlled robots are limited in their applications.

That is, for these robots to perform a different task, the task program as well as the control program must be adapted altogether to enable them to take on the new assignment. Although the number of commands in the task and control programs could be rather large and the level of detail in each command could be significantly great, there is no sequence of actions that these robots will alter in themselves in response to variation in data or operating conditions: Hence for each new task, the robots need to be reprogrammed, and this is time-consuming.

Data Processing

More so, as the number of assignments for such robots increases, and with each quite different from the other, the amount of data to be processed increases proportionately. For instance, mobile industrial robots like Automatic Guided Vehicles (AGVs), which are employed in warehouses to convey products from one place to another, have now in their development been equipped with a scanning laser sensor that triangulates their position by measuring reflections from fixed retro-reflectors. Therefore, as the number of obstacles to be maneuvered increases, so are the reflections and hence more data to be processed.

Furthermore, with many obstacles to be maneuvered, these industrial robots like AGVs are now equipped with machine vision, which requires the processing of enormous amounts of data. That can only be accomplished by high-speed digital computers. This is even more so in industrial tasks, such as part identification, quality inspection, and robot guidance. Another case study is the industrial mechanical robot arm—it is programmed to move through a sequence of motions in order to perform tasks, like loading and unloading of parts in a production line.

Limited Data Storage

Every instruction that computer-controlled robots must follow in order to accomplish a task must be clearly and explicitly spelled out in the task programs. Not only that, computer-controlled robots require a language processor to interpret the task programs and provide the data required by the control programs that direct the motions of the robots. This means that computer-controlled robots require large amounts of storage memory and are dependent on technological advancement in computer memory. Although the field of computer memory is rapidly advancing, it is still limited in what it can currently offer—this means that computer-controlled robots are largely limited in their capabilities.

Complexities in Programming

Moreover, assuming the act of programming is sophisticated enough that every detail of a command for the execution of a task can be programmed easily, there is still the problem of data storage since computer-controlled robots utilize computer data-storage technology to store program information. This means that the development of such robots is firmly dependent on improvements in computer data storage and software technologies. Not only that, the movement of computer-controlled robots is helplessly dependent on a feedback control system.

Limitation of Feedback Control

Feedback control system is the implementation of mathematical control theory that ensures the proper execution of the instructions in the task program in synergy with the control program in order for the robots to accomplish a task. For instance, during manufacturing, materials, and products are moved from one location to another in a series of processing steps; at each step, accurate positioning of such materials and products is indispensable, and it is the feedback control that confirms that the position has been moved accurately. In any case, the robots are to be made adaptable to continuously changing and unpredictable environments and the feedback must be implemented under such conditions. That is not a walk in the park.

Conclusion

Computer-controlled robots are monotonous. They are mostly able to perform a sequence of processing operations that are fixed by the equipment configuration and command. This makes them rather expensive and is mostly used in industries for productions that are made in large volumes. In a situation, there is the need to use such robots for an assignment so different, and assuming that they are reprogrammable, an operator has to take responsibility for re-instructing the robots since they lack the intelligence to re-instruct themselves. Therefore, computer-controlled robots are helplessly dependent on human operators in addition to complicated programming.