scholarly journals Vision-Based Robotic Arm Control Algorithm Using Deep Reinforcement Learning for Autonomous Objects Grasping

2021 ◽  
Vol 11 (17) ◽  
pp. 7917
Author(s):  
Hiba Sekkat ◽  
Smail Tigani ◽  
Rachid Saadane ◽  
Abdellah Chehri
Keyword(s):  
Reinforcement Learning ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Three Dimensional ◽  
Robotic Arm ◽  
Object Based ◽  
Policy Gradient ◽  
Intended Object ◽  
Autonomous Objects ◽  
Robotic Arm Control

While working side-by-side, humans and robots complete each other nowadays, and we may say that they work hand in hand. This study aims to evolve the grasping task by reaching the intended object based on deep reinforcement learning. Thereby, in this paper, we propose a deep deterministic policy gradient approach that can be applied to a numerous-degrees-of-freedom robotic arm towards autonomous objects grasping according to their classification and a given task. In this study, this approach is realized by a five-degrees-of-freedom robotic arm that reaches the targeted object using the inverse kinematics method. You Only Look Once v5 is employed for object detection, and backward projection is used to detect the three-dimensional position of the target. After computing the angles of the joints at the detected position by inverse kinematics, the robot’s arm is moved towards the target object’s emplacement thanks to the algorithm. Our approach provides a neural inverse kinematics solution that increases overall performance, and its simulation results reveal its advantages compared to the traditional one. The robot’s end grip joint can reach the targeted location by calculating the angle of every joint with an acceptable range of error. However, the accuracy of the angle and the posture are satisfied. Experiments reveal the performance of our proposal compared to the state-of-the-art approaches in vision-based grasp tasks. This is a new approach to grasp an object by referring to inverse kinematics. This method is not only easier than the standard one but is also more meaningful for multi-degrees of freedom robots.

Explicit inclusion of connectivity in geostatistical facies modelling.

2020 ◽  
Author(s):  
Tom Manzocchi ◽  
Deirdre Walsh ◽  
Carneiro Marcus ◽  
Javier López-Cabrera ◽  
Soni Kishan
Keyword(s):  
Statistical Physics ◽  
Degrees Of Freedom ◽  
Volume Fraction ◽  
Input Parameter ◽  
Three Dimensional ◽  
Training Image ◽  
Geostatistical Approach ◽  
Object Based ◽  
Facies Modelling ◽  
Facies Models

<p>Irrespective of the specific technique (variogram-based, object-based or training image-based) applied, geostatistical facies models usually use facies proportions as the constraining input parameter to be honoured in the output model. The three-dimensional interconnectivity of the facies bodies in these models increases as the facies proportion increases, and the universal percolation thresholds that define the onset of macroscopic connectivity in idealized statistical physics models define also the connectivity of these facies models. Put simply, the bodies are well connected when the model net:gross ratio exceeds about 30%, and because of the similar behaviour of different geostatistical approaches, some researchers have concluded that the same threshold applies to geological systems.</p><p>In this contribution we contend that connectivity in geological systems has more degrees of freedom than it does in conventional geostatistical facies models, and hence that geostatistical facies modelling should be constrained at input by a facies connectivity parameter as well as a facies proportion parameter. We have developed a method that decouples facies proportion from facies connectivity in the modelling process, and which allows systems to be generated in which both are defined independently at input. This so-called compression-based modelling approach applies the universal link between the connectivity and volume fraction in geostatistical modelling to first generate a model with the correct connectivity but incorrect volume fraction using a conventional geostatistical approach, and then applies a geometrical transform which scales the model to the correct facies proportions while maintaining the connectivity of the original model. The method is described and illustrated using examples representative of different geological systems. These include situations in which connectivity is both higher (e.g. fluid-driven injectite or karst networks) and lower (e.g. many depositional systems) than can be achieved in conventional geostatistical facies models.</p>

CNN architecture for robotic arm control in a 3D virtual environment by means of by means of EMG signals

2017 ◽  
Vol 10 (28) ◽  
pp. 1377-1390 ◽  
Author(s):  
Natalie Segura Velandia ◽  
Robinson Jimenez Moreno ◽  
Ruben Dario Hernandez
Keyword(s):  
Neural Network ◽  
Virtual Environment ◽  
Degrees Of Freedom ◽  
Wavelet Packet ◽  
Robotic Arm ◽  
Surface Electrodes ◽  
Average Accuracy ◽  
3D Virtual Environment ◽  
Robotic Arm Control ◽  
Electromyography Signals

This paper presents the development of a 3D virtual environment to validate the effectiveness of a Convolutional Neural Network (CNN) in a virtual application, controlling the movements of a manipulator or robotic arm through commands recognized by the network. The architecture of the CNN network was designed to recognize five (5) gestures by means of electromyography signals (EMGs) captured by surface electrodes located on the forearm and processed by the Wavelet Packet Transform (WPT). In addition to this, the environment consists of a manipulator of 3 degrees of freedom with a final effector type clamp and three objects to move from one place to another. Finally, the network reaches a degree of accuracy of 97.17% and the tests that were performed reached an average accuracy of 98.95%.

A complete analytical solution to the inverse kinematics of the Pioneer 2 robotic arm

Robotica ◽  
2005 ◽  
Vol 23 (1) ◽  
pp. 123-129 ◽  
Author(s):  
John Q. Gan ◽  
Eimei Oyama ◽  
Eric M. Rosales ◽  
Huosheng Hu
Keyword(s):  
Analytical Solution ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Robotic Manipulators ◽  
Robotic Arm ◽  
Effective Solution ◽  
Robotic Arms ◽  
Unstructured Environment ◽  
Inverse Kinematics Problem

For robotic manipulators that are redundant or with high degrees of freedom (dof), an analytical solution to the inverse kinematics is very difficult or impossible. Pioneer 2 robotic arm (P2Arm) is a recently developed and widely used 5-dof manipulator. There is no effective solution to its inverse kinematics to date. This paper presents a first complete analytical solution to the inverse kinematics of the P2Arm, which makes it possible to control the arm to any reachable position in an unstructured environment. The strategies developed in this paper could also be useful for solving the inverse kinematics problem of other types of robotic arms.

scholarly journals A comparison of three-dimensional rotation formalisms for least-squares and Bayesian inverse kinematics

2021 ◽  
Author(s):  
Ben Serrien ◽  
Klevis Aliaj ◽  
Todd Pataky
Keyword(s):  
Least Squares ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Estimation Error ◽  
Computational Cost ◽  
Three Dimensional ◽  
Linear Least Squares ◽  
Computationally Efficient ◽  
Rotational Degrees Of Freedom ◽  
Value Decomposition

Marker-based inverse kinematics (IK) is prone to errors arising from measurementnoise and soft-tissue artefacts. Various least-squares and Bayesian methods canbe applied to limit the estimation error to a minimum. Recently proposed meth-ods like Bayesian IK come at an increased computational cost however. In thistechnical paper, we present an overview of eight different least squares or BayesianIK methods, including their accuracy and computational load for IK problemsinvolving a single rigid body and three rotational degrees-of-freedom, whose at-titude is estimated from four noisy marker positions. The results indicate thatNon-Linear Least Squares, Variational Bayesian and full Bayesian IK are supe-rior to Singular Value Decomposition in terms of accuracy, with approximatelya two-fold error reduction. However, only Non-Linear Least Squares and Varia-tional Bayesian IK are computationally efficient enough to scale towards practicaluse in biomechanical applications, with computational durations of 1-10 ms; fullyBayesian procedures required approximately 30 s for single rotation calculations.All Python code and supplementary material can be found in this paper’s GitHubrepository: https://github.com/benserrien/pybik.

scholarly journals Design Of 4DOF 3D Robotic Arm to Separate the Objects Using a Camera

2021 ◽  
Vol 3 (1) ◽  
pp. 27-35
Author(s):  
Akhmad Fahruzi ◽  
Bimo Satyo Agomo ◽  
Yulianto Agung Prabowo
Keyword(s):  
Inverse Kinematics ◽  
Robotic Arm ◽  
Robot Arm ◽  
Fixed Position ◽  
End Effector ◽  
Robotic Arms ◽  
Work Area ◽  
Intended Object ◽  
Place Of Origin ◽  
Prototype Design

Nowadays robotic arm is widely used in various industries, especially those engaged in manufacturing. Robotic arms are usually used to perform jobs such as picking up and moving goods from their place of origin to the location desired by the operator. In this study, a 3d 4 DOF (Degree of Freedom) robotic arm. The prototype was made to move goods with random coordinates to places or boxes whose coordinates were determined in advance. The robot can know the coordinates of the object to be taken or moved. The arm robot prototype design is completed with a camera connected to a computer, where the camera is installed statically (fixed position) above the robot's work area. The camera functions like image processing to detect the object's position by taking the coordinates of the object. Then the object coordinates will be input into inverse kinematics that will produce an angle in every point of the servo arm so that the position of the end effector on the robot arm can be founded and reach the intended object. From the results of testing and analysis, it was found that the error in the webcam test to detect object coordinates was 2.58%, the error in the servo motion test was 12.68%, and the error in the inverse kinematics test was 7.85% on the x-axis, the error was 6.31% on the y-axis and an error of 12.77% on the z-axis. The reliability of the whole system is 66.66%.

scholarly journals Real-Time Posture Control for a Robotic Manipulator Using Natural Human–Computer Interaction

Proceedings ◽  
2018 ◽  
Vol 4 (1) ◽  
pp. 31
Author(s):  
Guillaume Plouffe ◽  
Pierre Payeur ◽  
Ana-Maria Cretu
Keyword(s):  
Real Time ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Robotic Arm ◽  
Human Hand ◽  
Graphical Interface ◽  
Left Hand ◽  
Human Control ◽  
Depth Data ◽  
Three Degrees Of Freedom

In this paper, we propose a vision-based recognition approach to control the posture of a robotic arm with three degrees of freedom (DOF) using static and dynamic human hand gestures. Two different methods are investigated to intuitively control a robotic arm posture in real-time using depth data collected by a Kinect sensor. In the first method, the user’s right index fingertip position is mapped to compute the inverse kinematics on the robot. Using the Forward And Backward Reaching Inverse Kinematics (FABRIK) algorithm, the inverse kinematics (IK) solutions are displayed in a graphical interface. Using this interface and his left hand, the user can intuitively browse and select a desired robotic arm posture. In the second method, the user’s left index position and direction are respectively used to determine the end-effector position and an attraction point position. The latter enables the control of the robotic arm posture. The performance of these real-time natural human control approaches is evaluated for precision and speed against static and dynamic obstacles.

Three-Dimensional Kinematic Analysis of a Two Actuated Spoke Wheel Robot Based on Its Equivalency to a Serial Manipulator

2008 ◽  
Author(s):  
Ping Ren ◽  
Ya Wang ◽  
Dennis Hong
Keyword(s):  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Three Dimensional ◽  
Forward Kinematics ◽  
Future Research ◽  
Ground Contact ◽  
Contact Points ◽  
Prismatic Joints ◽  
Forward Kinematic ◽  
The Relationship

In this paper, the inverse and forward kinematics of a novel mobile robot that utilizes two actuated spoke wheels is presented. Intelligent Mobility Platform with Active Spoke System (IMPASS) is a wheel-leg hybrid robot that can walk in unstructured environments by stretching in or out three independently actuated spokes of each wheel. First, the unique locomotion scheme of IMPASS is introduced. Then the configuration of the robot when each of its two spoke wheels has one spoke in contact with the ground is modeled as a two-branch parallel mechanism with spherical and prismatic joints. An equivalent serial manipulator of the 2-SP mechanism with the same degrees of freedom is proposed to solve for the inverse and forward kinematic problems. The relationship between the physical limits of the stroke of the spokes (effective spoke length) and the limits of its equivalent degree of freedom is established. This approach can also be expanded to deal with the forward and inverse kinematics of other configurations which has more than two ground contact points. Several examples are used to illustrate the method. The results obtained will be used in the future research on the motion planning of IMPASS walking in unstructured environment.

scholarly journals Robotic Obstacle Avoidance: A Virtual Modeling And Reinforcement Learning

2021 ◽  
Author(s):  
Ming-Fei Chen ◽  
Han-Hsien Tsai ◽  
Wen-Tse Hsiao
Keyword(s):  
Reinforcement Learning ◽  
Route Planning ◽  
Target Position ◽  
Learning System ◽  
Robotic Arm ◽  
Policy Gradient ◽  
Visualization Toolkit ◽  
Reward Functions ◽  
Virtual Modeling ◽  
Self Learning

Abstract This study developed a robotic arm self-learning system based on virtual modeling and reinforcement learning. Using the model of a robotic arm, information concerning obstacles in the environment, initial coordinates of the robotic arm, and the target position, this system automatically generated a set of rotational angles to enable a robotic arm to be positioned such that it can avoid all obstacles and reach a target. The developed program was divided into three parts. The first part involves robotic arm simulation and collision detection; specifically, images of a six-axis robotic arm and obstacles were input to the Visualization ToolKit library to visualize the movements and surrounding environment of the robotic arm. Subsequently, an oriented bounding box algorithm was used to determine whether collisions had occurred. The second part concerned machine-learning–based route planning. The TensorFlow was used to establish a deep deterministic policy gradient model, and reinforcement learning was employed for the response to environmental variables. Different reward functions were designed for tests and discussions, and the program’s practicality was verified through actual machine operations. Finally, the application of reinforcement learning in route planning for a robotic arm was proved feasible by the experiment. Such an application facilitated automatic route planning and achieved an error of less than 10 mm from the target position.

Another Approach for Redundancy Resolution of a 7 DOF Robotic Arm

2015 ◽  
Vol 762 ◽  
pp. 305-311
Author(s):  
Mihai Crenganis ◽  
Octavian Bologa
Keyword(s):  
Fuzzy Logic ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Robotic Arm ◽  
Cad Model ◽  
Redundancy Resolution ◽  
The Real ◽  
Kinematics Simulation ◽  
Inverse Kinematics Problem ◽  
Mathematical Equations

In this paper we have presented a method to solve the inverse kinematics problem of a redundant robotic arm with seven degrees of freedom and a human like workspace based on mathematical equations, Fuzzy Logic implementation and Simulink models. For better visualization of the kinematics simulation a CAD model that mimics the real robotic arm was created into SolidWorks® and then the CAD parts were converted into SimMechanics model.

Inverse Kinematics for a 7 DOF Robotic Arm Using the Redundancy Circle and ANFIS Models

2014 ◽  
Vol 657 ◽  
pp. 823-828
Author(s):  
Mihai Crenganis ◽  
Radu Eugen Breaz ◽  
Sever Gabriel Racz ◽  
Octavian Bologa
Keyword(s):  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Robotic Arm ◽  
Cad Model ◽  
The Real ◽  
Kinematics Simulation ◽  
Inverse Kinematics Problem ◽  
Mathematical Equations

In this paper we have presented a method to solve the inverse kinematics problem of a redundant robotic arm with seven degrees of freedom and a human like workspace based on mathematical equations, ANFIS implementation and Simulink models. For better visualization of the kinematics simulation a CAD model that mimics the real robotic arm was created into SolidWorks® and then the CAD parts were converted into SimMechanics model.

Sign in / Sign up

Export Citation Format

Share Document