Cristian Pop

Geared Linkages with Linear Actuation Used as Kinematic Chains of a Planar Parallel Manipulator

The paper deals with a novel kinematic chain, which uses a geared linkage with linear actuation. The structure is used to actuate the mobile platform of a parallel planar manipulator (3-RRR). The main characteristics of the geared linkage with linear actuation are the large rotation angle with proper transmission angle and the approximately linear transmission function in a large range. These properties of the geared linkages with linear actuation as kinematic chain allow the avoiding of the first type singularities and an easier control of the mobile platform movement.

Colored object detection algorithm for visual-servoing application

A visual servoing application for automated object detection is described. This application is based on a computer vision algorithm that enables: the detection of colored objects, which are located on a conveyor pallet, the height calculation of these objects, the determination of their type, their position and orientation, in order for a SCARA robot to be able to manipulate them. The object detection algorithm developed with the help of Matlab software relies on the fact that the image segmentation operation is performed in the HSV color space. The camera calibration and the robot calibration methods used are also described.

Novel Solution for Leg Motion with 5-Link Belt Mechanism

Abstract From the analysis of Theo Jansen walking mechanism and of the path curve that it describes the reduced capability for crossing over obstacles of the Jansen leg (1 DOF) is pointed out. By using a 5 link belt mechanism with 2 DOF can be adapted for generating similar Jansen mechanism path curve, where the step height of this path can be increased. For this purpose a mathematical model is conceived in order to analyse and determine the parameters for driving and control of the operation of the novel walking leg solution.

Image Processing and Artificial Neural Network for Robot Application

This paper presents a robot vision application, implemented in MATLAB working environment, developed for feature-based object recognition, object sorting and manipulation, based on shape classification and its pose calculus for proper positioning. The application described in this article, designed to detect, identify, classify and manipulate objects is based on previous robot vision applications that are presented in more detail in [1]. The idea underlying the mentioned applications is to determine the type, position and orientation of the work pieces (in those cases different types of bearings). Taking it further, in the presented application, objects that show shape with a gradual level of complexity are used. For this reason pattern recognition are discriminated by training a two layers neural network. The network is presented and also the input and output vectors.

Trajectory Analysis for Modified Jansen Leg Mechanism Configuration

Starting from a previous research regarding Jansen type leg mechanism that was implemented in the structure of a walking robot, few problems were identified. One of these problems was related to the reduced number of DOFs that it possesses and the step height described by the end point of the leg during walking. After conducting a series of simulations regarding the path curves described by the end point of Jansen’s leg configuration, a limited number of paths are analysed. In spite of the numerous possibilities for varying the link lengths, it has been observed that for a slight variation of the distance between the fixed joints along horizontal axis, the step height can be increased. The shapes of the path curves described by these modified configurations are analysed and compared with the original one based on the length and the height of the step values. Based on the observations made upon the new path curves obtained a practical solution is proposed. This consists in adding a second DOF to the leg mechanism that will allow adjusting the distance between the fixed joints.

Dimensional synthesis of a leg mechanism

An eight bar leg mechanism dimensional synthesis is presented. The mathematical model regarding the synthesis is described and the results obtained after computation are verified with help of 2D mechanism simulation in Matlab. This mechanism, inspired from proposed solution of Theo Jansen, is integrated into the structure of a 2 DOF quadruped robot. With help of the kinematic synthesis method described, it is tried to determine new dimensions for the mechanism, based on a set of initial conditions. These are established by taking into account the movement of the end point of the leg mechanism, which enters in contact with the ground, during walking. An optimization process based on the results obtained can be conducted further in order to find a better solution for the leg mechanism.

A Dynamic Analysis Based on MBD ADAMS Program for a Variant of Quadruped Robot

For stability and impact reaction forces assessment of a quadruped robot during walking, a dynamic analysis is considered. For this purpose, a variant of a quadruped robot based on Jansen mechanism is presented. For interpreting the influence of the reaction forces from the ground during walking, the analysis was conducted with help of ADAMS software using a 3D model of the robot. Material specifications, forces and moments acting in the robot structure were considered. Graphical results obtained regarding the ground reaction forces are displayed. Also a reduced mass moment of inertia at the crankshaft is taken into consideration based on Lagrange motion equation and generalized coordinates.

Analysis of the Trajectory Shape Described by a Robotic Leg during a Walking Sequence

An experimental test bench is proposed in order to analyze the trajectory shape described by the termination of the Jansen’s type of leg mechanism during a walking sequence. A prototype of a walking robot based upon Jansen walking mechanism is constructed for this purpose. A marker is placed at the end of the leg and a CCD camera recorded the trajectory shape described by the leg driven by a DC motor through a crank. The results are interpreted with the help of a program based on image processing created for determining the length and the height of the step during walking after studying the trajectory.

Robot Vision Application for Overlapped Work Pieces

An automated robot vision application for work piece manipulation is described. The purpose is to pick up work pieces like bearings from a pile and store them in a specific location. The necessary image processing algorithms developed for this application and the camera calibration process are realized by using Matlab software. Based on these algorithms the operation for detecting, identifying, sorting and manipulating the bearings is realized. Also the mathematical model for determining the position and orientation of the bearing with respect to the world reference system attached to the robot’s base is presented.

Synthesis of an articulated ten bar mechanism for building a walking robot

A ten bar walking mechanism synthesis is presented. The method stated in this case is based on Freudestein equations which are frequently used for bar mechanism synthesis. Following the results of synthesis a CAD model has been achieved and a prototype of a mobile walking robot is described.