Defined as a mix of electronics, mechanics and programming, a robot is supposed to “sense” information from its environment through its sensors, analyse and deal with the sensor data, come to a decision and carry out necessary actions or perform something in response, regardless of the robot is autonomous, or guided by a human.
Most of the robots are semi-autonomous, i.e. some decisions are intelligently controlled by the robot, while others are controlled by a human. For example, the case of Mars Rover, which was able to operate with own intelligence while taking high-level …show more content…
The common control technique is to supply instructions to a robot so that it responds in a specific way for specific situations. For example, for an autonomous robot that avoids objects, the application tells the robot to back off anytime the robot comes into an object: a specific response to a specific condition. This activity does not need any human guidance, it is pre-programmed and does not give intelligence to a robot.
Fuzzy logic, evolutionary computation, hierarchical control, neural networks, etc. are some of the control techniques used for autonomous robots. Neural networks help a robot learn from its experience and environment and act specifically. Research in artificial intelligence is evolving, but for the time being, the robot does what it is set to do.
Rich knowledge in computer science, mathematics, electrical and mechanical engineering, systems engineering, is required to grasp the complexity of robots and their …show more content…
Considering the three-layer architecture approach, the lowest level of control is the behavioral control (driver). It directly connects sensors and actuators. While these are typically hand-crafted functions written in C or C++, there have been specialized languages developed for behavioral control, including ALFA, Behavioral Language and REX. It is at this level that traditional control theory resides [1].
The second layer is the executive layer (platform). It represents the interface between the numerical behavioral control and the symbolic planning layers and it is responsible for translating high-level plans into low-level behaviors, invoking behaviors at the appropriate times, monitoring execution, and handling