Free Robot ARM Essay Example

A robotic arm is a kind of mechanical arm, mostly programmable; that mimics the functions of a human arm. The close emulation of a human hand is desired in robotic arms because it increases the number of applications for which it can be used. A typical arm consists of 3-7 metal segments connected by 3-6 joints. A computer controls the robotic arm by rotating the servo motors linked to each joint. The robotic arm also uses sensors to control the movement of the parts to the required extent. Sensors are used to determine the system’s current state. Robotic applications call for sensors with a high degree of repeatability, reliability and precision. Sensors are connected to motors and arms in various ways depending on the type of sensor and how it operates. They are connected to individual step motors that move in exact increments, enabling the arm to move with great precision.
Position sensors (in the form of radial potentiometers) are connected to the motors through a gear mechanism. As the gears rotate, they change the resistance by varying the position of a pointer along the radial potentiometer sensor. The change in resistance in the sensor produces an equivalent voltage change from the sensor/ potentiometer. A signal modulated through pulse width is sent through the control wire. The pulse width is converted to an equivalent voltage that is compared to that of the potentiometer/sensor using an error amplifier.
A tactile sensor is another type of sensor that may be used in arms and motors. Tactile sensors are connected to motors and arms through a microswitch that has a lever arm that is activated to press a switch nub. This arrangement may be used when an arm or motor reaches a given limit or position. The activation of the switch provides feedback on the position of the arm or motor.
Figure 1: Sensor connected through a microswitch/ Limit Switch

An ultrasonic distance sensor uses sonar technology to establish the radial distance between the arm and an object.

Sensors are powered depending on their stipulated operating power. The average potentiometer sensor uses 5 volts power and is connected to the +5V terminals and the ground terminal.
Figure 2: Sensor connected to a divider circuit
Most sensors are connected to robotic circuits through voltage divider circuits as shown in fig 2. In most applications, the R1 resistor has a constant or fixed value as shown in figure 3. R2 is the sensor’s variable resistance. Vin is the voltage supply (5 Volts positive). The Vout signal may be calculated directly from R2, the sensor. From the equation Vout = Vin (R2 / R1+R2), If R2 is large relative to R1, the output voltage is large. If R2 is small relative to R1, there is a small output voltage. The voltage range varies between minimum and maximum voltage values of 0 and 5 Volts. The circuitry in figure 3 is used for every individual sensor input channel. The most important element of this circuit is the pull-up resistor connected from the input signal lead of the sensor to the 5 volt supply of power. The pull-up resistor is used for two reasons. First, it gives a default value for the input of the sensor. The default value is the value when no sensor is connected. Many integrated circuits (ICs) do not perform well if their inputs are left unconnected.
Figure 3: Sensor Input Port Circuitry
Figure 4: Sample flow diagram showing power supply to servos, microprocessor, sensor, etc
Sensors provide feedback signals to the circuit board regarding the position of the arm or motor. The circuit board electronics decode the signals provided by the sensor to determine the required position of the arm. The electronics or microcontrollers further compare the current position of the arm or motor to the desired position and determines which direction to turn the shaft to acquire that desired position. This work is performed by the microcontroller circuit, which serves as the brain of the robotic arm.
The signals sent to the microcontroller are referenced to previously recorded positions of the arm. This process happens in a process known as “lead-through” or “record playback.” Then desired positions are pre-recorded in memory in a step-by-step fashion. In playback mode, an operator observes the sequence and modifies it accordingly. With more challenging applications, some jobs call for continuous path control of end effectors. All actions are programmed in appropriate speeds for particular tasks. Controllers are run through the code. Codes set the limits of the robot and what it can do.
Sensors give feedback continuously through a process known as feedback (closed-loop) control. Feedback control is a method of getting a system (the robotic arm in this case) to achieve and maintain a particular desired state through a continuous process of comparing its current state to its desired state. The desired state is referred to as the system’s goal state. This state may be internal or external. In the case of a robotic arm, the goal state could be a given position or angle or distance from an object. If the current state equals the desired state, there is no need for the system to do anything. However, if the states are not equal, the action taken by the system depends on its design. The difference between current and desired states is referred to as the error. The goal of a system is to reduce the error as much as possible. Sensors provide constant feedback providing the magnitude and direction of the error.
Sensor control then becomes the calculation of the path desired. This process may be achieved by using the deviations between actual and desired paths to calibrate the sensor. The advantage of this approach is that a new sensor may be connected without having to be tuned. Even when large time delays occur, no oscillations occur. One challenge of direct sensor feedback has been referred to as a “space-time problem.” This means that when a path deviation is sensed, there is a time-lag until the system undertakes counter-measures. This delay is attributable to robot dynamics and sensor processing.
The other type of feedback that makes use of sensors is known as reactive control. Reactive control is based on a tight loop that links the sensors of the robot to its effectors. Purely reactive systems react to current sensory information. They use a direct mapping between sensors and effectors. They comprise of collections of rules that map given situations to specific actions. Reactive control is a recent advancement in robotics that helps to avoid errors. For example, a robotic arm that misses the intended position of welding may use other sensors such as camera sensors and color sensors to locate the target and correct its trajectory.