Kanny Krizzy D. Serrano

Development of a variable negative pressure Jamming Gripper through visual object size classification and Artificial Neural Network

The Universal Jamming Gripper, made up of a coffee grain-filled balloon membrane and a vacuum pump, operates by applying an unvarying or constant negative pressure set by the manipulator on the target object for gripping, not taking into account the suitable amount of negative pressure depending on the size and weight of the target object. With the constant or same amount of negative pressure applied on the target objects of different sizes and weight, the gripper can diminish the structural integrity of some of these target objects. To eradicate this problem, the researchers developed a system to vary the negative pressure of the Universal Jamming Gripper through an artificial neural network depending on the size of the object, obtained through a vision-based object classification scheme, and weight obtained from a load cell connected to the computer. An Artificial Neural Network (ANN), trained with three inputs, such as the pixel area of one side of the target object, the pixel area of another side of the object and its weight, is used to automatically determine the optimum negative pressure needed to successfully grip the target object. After testing and experimentation, the ANN is proven to output the optimum negative pressure needed to successfully conform to and grip the target object, as evident with the 99.131% result from testing, based on the regression plot from MatLab.

An improved collision avoidance scheme using Artificial Potential Field with fuzzy logic

The Artificial Potential Field (APF), reputed for being one of the most prominent algorithm for traversing a path in the shortest distance possible, has a critical disability when confronted by obstacles classified by the state of being under local minima. The said state occurs when the APF algorithm yields an appreciable attractive force between the immediate position of the robot and the goal distance while, at the same time, yielding a considerable repulsive force due to the presence of obstacles that prevents it from determining its next position. The solution manifested by the researchers utilizes another reputed algorithm known as the Fuzzy Logic Inference (FLI) System by using membership functions that takes input from ultrasonic distance sensors and outputs left and right motor speed that implicitly controls the robot turn as well. Membership functions have a given pre-determined range of values that control the state of an input or an output. The fuzzy rule base, a linguistic set of IF-THEN statements, used in this FLI system follows a wall-following robot behaviour accessed if and only if the robot determines that it is under the state of local minima. Simulation and experiments of their algorithm shows good performance and ability to overcome the local minimum problem associated with potential field methods.

Autonomous trajectory tracking of a quadrotor UAV using PID controller

Quadrotor helicopters are becoming popular rotorcrafts for unmanned aerial vehicles (UAV) platforms because of its simplicity, high maneuverability and the ability to withstand harsh environments without the risk of human intervention. This now imposes a demand in areas of control and navigation. For the quadrotor to follow a desired path, a controller must be able to track its trajectory and ensure that the error is minimal with the presence of external disturbances. The study will utilize the use of a PID as the controller of the Quadrotor UAV. PID controller is popular because it has good control performance and is easy to implement but the changing of its coefficients are performed manually.

Machine vision system, robot hardware design and fuzzy controller design for autonomous multi-agent mobile robot platform

In this paper we present the machine vision system and the hardware design of our mobile robot with the design of the fuzzy controller. This paper describes the implementation of robots position and tracking with a use of a vision system. The ground truth data is obtained by the overhead camera over the setup to track the current position of the robot. The process in the camera includes robot detection, image to coordinate transformation, and background subtraction. Several positions were chosen to test the vision system and tracking of the mobile robot using Roborealm. The hardware design of the mobile robot is presented using Sketch Up Pro. The overall dimension was limited to 7.5 cm × 7.5 cm × 7.5 cm in compliance with the standard of the Federation of International Robot-Soccer Association for Micro-soccer Robot World Tournament.