Color Sorting Robotic Arm with Servo Motors and Arduino

This tutorial demonstrates how to detect shapes (rectangle, triangle, circle) and colors in real-time video using OpenCV and Python by converting frames to HSV color space, creating binary masks with trackbars for adjustable thresholds, finding contours, filtering by area, approximating polygon vertices, and classifying shapes based on vertex count.

This tutorial covers the fundamentals of image processing using OpenCV in Python, including reading and displaying images with imread() and imshow(), drawing geometric shapes (lines, rectangles, circles, ellipses, polygons), performing arithmetic operations (addition, subtraction, bitwise operations), applying edge detection, scaling and rotating images, creating image borders, playing videos, capturing camera feed, implementing morphological operations, working with image pyramids, translating images, applying color space conversions, cropping images, creating blank images, using trackbars for interactive controls, selecting regions of interest, flipping and transposing images, saving processed videos, analyzing histograms, performing thresholding (simple and Otsu's), detecting contours and their moments, implementing Hough transformations for lines and circles, using template matching, applying GrabCut for background removal, and detecting objects like faces, eyes, smiles, and pedestrians using Haar cascades and MediaPipe.

Template matching locates occurrences of a template shape within a larger image. Convert both images to grayscale. Use matchTemplate() with the template and scene images. The result contains scores indicating match quality. Use minMaxLoc() to find the best match location. Access the best match index from the results vector to locate and draw the detected shape.

This tutorial demonstrates how to recognize geometric shapes (rectangle, circle, square, triangle) in images using OpenCV Python through a systematic process: converting color images to grayscale, applying thresholding to create binary images, extracting contours, approximating contours with polygons to count vertices, and classifying shapes based on vertex count and aspect ratio (width/height ratio between 0.95-1.05 indicates a square).

A closed-loop control system for robotic joints uses feedback to achieve accurate position control. The architecture includes: (1) A summing junction comparing desired position θ_d with actual position θ to compute error e, (2) A derivative block computing error rate ė, (3) Gain blocks applying KP, KI, and KD to their respective signals, (4) A summer combining all terms to generate τ', and (5) A plant (motor and mechanical system) producing actual output θ. Sensors like optical encoders measure θ and θ̇ for feedback. This architecture enables the robot to adjust its behavior continuously based on actual performance versus desired trajectory.

Robotic manipulators demonstrate closed-loop control principles with continuous feedback. The system includes: Sensor (camera/encoder measuring arm position), Controller (microprocessor/microcontroller), Power Source, Actuators (motors, grippers), and Plant (robotic arm). Controlled variables include position, velocity, acceleration, torque, and force. The feedback mechanism works through continuous sensing and comparison: the controller receives position measurements, compares them against the reference position, and generates commands to adjust actuators until the desired position is achieved. This enables precise object manipulation tasks.

Multiple practical examples demonstrate closed-loop control systems: (1) Motor speed control uses a potentiometer for reference, differential amplifier for comparison, and tachogenerator as sensor to measure angular velocity; (2) Automotive direction control uses the driver as both control element and sensor, comparing desired path (road markings) with actual direction through visual and tactile feedback; (3) Robotic manipulator position control uses camera feedback to measure end-effector position and compares it with desired trajectory, adjusting joint motors accordingly; (4) Digital temperature control uses analog-to-digital conversion, computer processing, and relay control to maintain desired temperature.

The complete feedback control loop consists of: (1) acquiring joint angles via get angles function, (2) comparing with setpoint, (3) processing through PID controller, (4) generating motor power commands, and (5) applying to motors. This closed-loop system continuously adjusts motor power based on the error between desired and actual joint angles.

A PID controller in closed-loop position control uses three components—proportional, integral, and derivative amplifiers—to precisely move a robotic arm to a desired position by comparing a command signal with a feedback signal from a potentiometer; the proportional amplifier responds to immediate error, the derivative amplifier predicts overshoot by responding to changing error rates, and the integral amplifier eliminates steady-state error caused by friction and mass, working together to achieve fast response with minimal overshoot.

Industrial automation integrates various devices including PLC controllers, pneumatic cylinders, sensors, and conveyors to streamline manufacturing processes; pneumatic cylinders use compressed air to perform linear motion tasks like sorting materials by color, while conveyors (such as slat and ruler conveyors) transport materials between workstations, with the entire system often controlled by PLCs communicating via protocols like OPC UA and MQTT to enable Industry 4.0 smart factory operations.

This segment introduces Programmable Logic Controllers (PLCs) as industrial computers for automating electromechanical processes. The video explains three automation levels: manual pneumatic operation, semi-automatic electro-pneumatic control, and fully automatic PLC-based operation. The PLC hardware (Fanuc IC200U-D104) operates on 24V DC with 9 pulse inputs (red=positive, black=negative). The pneumatic trainer uses two double-acting cylinders with solenoid valves to execute sequential operations: A+ (extend), B+ (extend), B- (retract), A- (retract). Two programming methods exist: ladder logic (graphical) and mnemonics (text-based), with the lab focusing on ladder logic for its intuitive relay-like appearance.

Industrial automation integrates continuous processes (like liquid brewing) with manufacturing processes (like bottling). PLC programming uses ladder logic with LogixPro software, featuring start/stop circuits with sealing contacts, conveyor belt pulse control using timers, and safety interlocks requiring multiple sensors to confirm safe conditions before allowing conveyor movement. These fundamentals form the basis for automated production line control.

This video demonstrates a systematic five-step methodology for programming PLCs to control pneumatic manufacturing processes: (1) define inputs (sensor S1) and outputs (pistons P1, P2, P3 and conveyor B1), (2) use timing diagrams to represent logical relationships, (3) translate these relationships into equations using cascading timers (Timer 1 activates Timer 2, which activates Timer 3, which activates Timer 4), (4) implement the logic in ladder programming language, and (5) apply the solution to specific PLC platforms (S7-200 and Logo 8). The practical application involves a manufacturing sequence where sensor S1 detects a product, triggering pistons P1 and P3 to position the piece for 2 seconds, then piston P2 to stamp for 2 seconds, followed by retraction of all pistons and conveyor restart.

This segment covers the complete PLC programming workflow for creating industrial automation systems. It begins with defining system requirements (conveyor motor control based on sensor detection), progresses through tag definition in TIA Portal for PLC variables, and culminates in writing simple ladder logic programs that implement basic control logic. The process demonstrates how to create a functional PLC program that responds to sensor inputs by controlling output devices like motors, forming the foundation for more complex automation systems.