Virtual Presentation - Adaptive Artificial Potential Fields with Orientation Control Applied to Robotic Manipulators

Authors: Caio Viturino, Ubiratan Junior, André Scolari, Leizer Schnitman

21st International Federation of Automatic Control World Congress (IFAC 2020)

Presentation language: English

Virtual presentation give at the International Federation of Automation Control 2020. This presentation is related to the published paper titled "Adaptive Artificial Potential Fields with Orientation Control Applied to Robotic Manipulators".

Video

Virtual Presentation - Convolutional Neural Networks applied in Object Identification and Robotic Grasping

Authors: Caio Viturino, Kleber Santana, Daniel Oliveira, Cézar Lemos, André Scolari

XXIII Congresso Brasileiro de Automática (CBA 2020)

Presentation language: Portuguese

Virtual presentation given at the XXIII Congresso Brasileiro de Automática (CBA) in 2020. This presentation is related to the published paper titled "Redes Neurais Convolucionais para Identificação e ​Preensão Robótica de Objetos​".

Video

Convolutional Neural Networks applied in Object Identification and Robotic Grasping

Authors: Caio Viturino, Kleber Santana, Daniel Oliveira, Cézar Lemos, André Scolari

XXIII Congresso Brasileiro de Automática (CBA 2020)

Paper language: Portuguese

We implemented a two-step cascaded system with the Generative Grasping Convolutional Neural Network (GG-CNN) and a modified version of the Single Shot Multibox Detector architecture (SSD) to perform robotic grasping in a vision-based object recognition system. The proposed method is called Single Shot Generative Grasping Neural Network (SSGG-CNN). The GG-CNN is a powerful object-independent grasping synthesis method well-known for the outstanding performance in open-loop and closed-loop systems using a pixel-wise grasp quality prediction. However, this technique does not allow the robot to selectively grasp objects. In order to mitigate this problem, the modified SSD was adopted to perform the object detection and selection preceding the grasping. It was achieved an average success rate of 85% considering open-loop and uncluttered objects randomly organized on a planar surface.

Code Video Paper

Adaptive Artificial Potential Fields with Orientation Control Applied to Robotic Manipulators

Authors: Caio Viturino, Ubiratan Junior, André Scolari, Leizer Schnitman

21st International Federation of Automatic Control World Congress (IFAC 2020)

Paper language: English

In this work, we proposed the integration of an Adaptive Artificial Potential Fields algorithm with a new end effector orientation control technique for real-time robot path planning. The development of autonomous robotic systems has undergone several advances in path planning algorithms. These systems generate object collision-free paths in the robot's workspace. In this context, the Artificial Potential Fields technique has been the focus of improvements in recent years due to its simplicity of application and efficiency in real-time systems, since it does not require a global mapping of the robot's workspace. In spite of its efficiency, this technique is susceptible to local minimum problems of different natures, such as Goals Non-Reachable with Obstacles Nearby (GNRON). To solve this problem, we suggest the use of an improvement called Adaptive Artificial Potential Fields used in conjunction with the proposed end effector orientation control technique, which allows reaching a desired orientation of the end effector. The resulting force, generated from the Adaptive Artificial Potential Field, guides the robot end effector to the goal. The Robot Operating System (ROS) framework and a collaborative robot manipulator UR5 are used to validate the proposed method on an approaching task for an object on a 3D printer tray.

Code Video Paper

Anti-collision System Applied To Robotic Manipulators Based on Artificial Potential Field Algorithm

Authors: Caio Viturino, Ubiratan Junior, André Scolari, Leizer Schnitman

Anais do 14º Simpósio Brasileiro de Automação Inteligente (SBAI 2019)

Paper language: Portuguese

We applied a path planning technique based on the Artificial Potential Fields on a robotic manipulator. This technique has been the focus of improvements in recent years due to its simplicity of application and efficiency in real-time systems, due to the fact that it does not require a global mapping of the robot's workspace. In spite of its efficiency, this technique is susceptible to local minima problems of different natures, such as: Goals Non-Reachable with Obstacles Nearby and Reacharound Local Minimum Problem. To solve these problems, an improvement called Adaptive Artificial Potential Fields is used in conjunction with the Subgoal Selection, Goal Configuration Sampling and Convex Hull techniques. The Robot Operating System (ROS) framework and a collaborative robot manipulator UR5 validate the proposed method.

Video Paper

Development of a 3 DOF Robotic Manipulator for educational purposes

Authors: Caio Viturino, Kleber Santana, Anderson Ribeiro Cesarino, Igor Cordeiro, Leizer Schnitman

XI Congresso de Engenharia, Ciência e Tecnologia (CONECTE 2018)

Paper language: Portuguese

In this work, we presented the development of a SCARA robot, with three degrees of freedom, for educational purposes. All robot parts were designed using SOLIDWORKS software, as well as deflection tests and static analysis. The links and base of the robot were built on a CNC machine, using aluminum 5052 H34 as material, and the gears were produced through 3D printing, using ABS. An algorithm capable of converting trajectories in the configuration space into a pulse train was developed to control the joint movement. For simulation and control, we used the Matlab/Simulink software and the Arduino Mega 2560 microcontroller. Experimental tests proved an accuracy of 1.5mm and repeatability of 2mm.

Paper Photo 1 Photo 2