Trajectory Prediction with a Conditional Variational Autoencoder

2019 ◽  
Vol 31 (3) ◽  
pp. 493-499
Author(s):  
Thibault Barbié ◽  
◽  
Takaki Nishio ◽  
Takeshi Nishida
Keyword(s):  
Motion Planning ◽  
Degrees Of Freedom ◽  
Industrial Robot ◽  
Computation Time ◽  
Prediction Algorithm ◽  
Robot Arm ◽  
Trajectory Prediction ◽  
Six Degrees Of Freedom ◽  
Variational Autoencoder ◽  
Planning Approaches

Conventional motion planners do not rely on previous experience when presented with a new problem. Trajectory prediction algorithms solve this problem using a pre-existing dataset at runtime. We propose instead using a conditional variational autoencoder (CVAE) to learn the distribution of the motion dataset and hence to generate trajectories for use as priors within the traditional motion planning approaches. We demonstrate, through simulations and by using an industrial robot arm with six degrees of freedom, that our trajectory prediction algorithm generates more collision-free trajectories compared to the linear initialization, and reduces the computation time of optimization-based planners.

Application of non-linear H∞ control to the Tetrabot robot manipulator

1998 ◽  
Vol 212 (6) ◽  
pp. 459-465 ◽  
Author(s):  
I Postlethwaite ◽  
A Bartoszewicz
Keyword(s):  
Degrees Of Freedom ◽  
Robot Manipulator ◽  
Industrial Robot ◽  
External Disturbances ◽  
Six Degrees Of Freedom ◽  
Control Scheme ◽  
Non Linear ◽  
Modelling Uncertainties ◽  
Real Robot ◽  
Robotic Application

In this paper, an application of a non-linear H∞ control law for an industrial robot manipulator is presented. Control of the manipulator motion is formulated into a non-linear H∞ optimization problem, namely optimal tracking performance in the presence of modelling uncertainties and external disturbances. Analytical solutions for this problem are implemented on a real robot. The robot under consideration is the six-degrees-of-freedom GEC Tetrabot. Investigations are made into the selection of weights for the H∞ controller and it is shown how different selections of weights affect the Tetrabot performance. The authors believe this to be the first robotic application of nonlinear H∞ control. Comparisons of the proposed control strategy with conventional proportional-derivative and proportional-integral-derivative controllers show favourable performance of the Tetrabot under the new non-linear H∞ control scheme.

scholarly journals Crack Assessment of Wheel Hubs via an Ultrasonic Transducer and Industrial Robot

Sensors ◽  
2018 ◽  
Vol 18 (12) ◽  
pp. 4336 ◽  
Author(s):  
Hanming Zhang ◽  
Chunguang Xu ◽  
Dingguo Xiao
Keyword(s):  
Crack Growth ◽  
Ultrasonic Testing ◽  
Degrees Of Freedom ◽  
Industrial Robot ◽  
Ultrasonic Transducer ◽  
Robot Assisted ◽  
Six Degrees Of Freedom ◽  
Crack Location ◽  
Critical Areas ◽  
Growth Assessment

Crack assessment when making fitness-for-service decisions requires a thorough examination of crack location and size in critical areas. An ultrasonic transducer is used for such assessments, but traditional methods cannot cope with complex rotators, such as wheel hubs. We present a model of robot-assisted crack growth assessment in wheel hubs. We integrate a six-degrees-of-freedom (DOF) industrial robot and a turntable to form a robot-assisted ultrasonic testing (UT) system that does not use traditional UT equipment. Ultrasonic beams are focused at certain depths appropriate for achieving maximum sensitivity. We quantitatively analysed wheel hubs with longitudinal and transverse series of pre-cracks, and concluded that our system autonomously detected cracks.

Real-time motion planning for robot manipulators in unknown environments using infrared sensors

Robotica ◽  
2007 ◽  
Vol 25 (2) ◽  
pp. 201-211 ◽  
Author(s):  
Shuguo Wang ◽  
Jin Bao ◽  
Yili Fu
Keyword(s):  
Motion Planning ◽  
Real Time ◽  
Collision Detection ◽  
Detection Method ◽  
Computation Time ◽  
Robot Arm ◽  
Infrared Sensors ◽  
Geometry Model ◽  
Time Motion ◽  
Series Of Experiments

SUMMARYThis paper deals with sensor-based motion planning method for a robot arm manipulator operating among unknown obstacles of arbitrary shape. It can be applied to online collision avoidance with no prior knowledge of the obstacles. Infrared sensors are used to build a description of the robot's surroundings. This approach is based on the configuration space but the construction of the C-obstacle surface is avoided. The point automation is confined on some planes with square grids in the C-space. A path-searching algorithm based on square grids is used to guide the automation maneuvering around the C-obstacles on the selected planes. To avoid the construction of the C-obstacle surface, the robot geometry model is expanded, and the static collision detection method is used. Hence, the computation time is reduced and the algorithm can work in real time. The effectiveness of the proposed method is verified by a series of experiments.

scholarly journals ABB IRB 120-30.6 Build Procedure in RoboDK

2021 ◽  
pp. 256-264
Author(s):  
Sudip Chakraborty ◽  
P. S. Aithal
Keyword(s):  
Design Methodology ◽  
Degrees Of Freedom ◽  
Industrial Robot ◽  
Industrial Robots ◽  
Research Purpose ◽  
Six Degrees Of Freedom ◽  
Simulation Based ◽  
Software Download ◽  
Coordinate Assignment

Purpose: Research on robotics needs a robot to experiment on it. The actual industrial robot is costly. So, the only resort is to use a Robot simulator. The RoboDK is one of the best robot simulators now. It has covered most of the popular industrial robots. Its interface is straightforward. Just open the software, download the robot as we need, and start experiments. Up to that, no issue was found anywhere. However, the problem begins when we want to build the simulated robot by own. Lots of complexity arises like coordinate assignment, rotation not aligned, length mismatch, robot not synced with DH parameter. We begin to find some documents for making the robots. A few bits of the document are present. That is why we research it. After doing that, we prepared this paper for the researcher who wants to develop the simulated robot independently. This paper can be referenced for them. To minimize the complexity of our research, we study an industrial robot, ABB IRB 120-30.6. It is a good and popular robot. It is six degrees of freedom robot. We will use the specification and STEP file from their respective website and build a simulated robot from the STEP file for our research purpose. Design/Methodology/Approach: We will create a simulated robot from ABB IRB 120-30.6 STEP file. To create a robot by own, we took the help of the IRB 120 robot model. To demonstrate as simple as possible, we start with that robot whose default design is already present. We match and tune the joint coordinate based on robot parameters through this experiment. Findings/results: Here, we see how to create a custom robot. Using the IRB 120 robot model, we will create a robot model step by step. Furthermore, it will move it around its axis. Originality/Value: Using this experiment, the new researcher can get valuable information to create their custom robot. Paper Type: Simulation-based Research.

scholarly journals Output Error Methods for Robot Identification

2019 ◽  
Vol 142 (3) ◽  
Author(s):  
Mathieu Brunot ◽  
Alexandre Janot ◽  
Francisco Carrillo ◽  
Joono Cheong ◽  
Jean-Philippe Noël
Keyword(s):  
Degrees Of Freedom ◽  
Target Identification ◽  
Industrial Robot ◽  
Computational Time ◽  
Dynamic Identification ◽  
Inverse Dynamic ◽  
Six Degrees Of Freedom ◽  
Output Error ◽  
Modeling Errors ◽  
Regression Techniques

Abstract Industrial robot identification is usually based on the inverse dynamic identification model (IDIM) that comes from Newton's laws and has the advantage of being linear with respect to the parameters. Building the IDIM from the measurement signals allows the use of linear regression techniques like the least-squares (LS) or the instrumental variable (IV) for instance. Nonetheless, this involves a careful preprocessing to deal with sensor noise. An alternative in system identification is to consider an output error approach where the model's parameters are iteratively tuned in order to match the simulated model's output and the measured system's output. This paper proposes an extensive comparison of three different output error approaches in the context of robot identification. One of the main outcomes of this work is to show that choosing the input torque as target identification signal instead of the output position may lead to a gain in robustness versus modeling errors and noise and in computational time. Theoretical developments are illustrated on a six degrees-of-freedom rigid robot.

Theory and Experiments on the Compliance Control of Redundant Robot Manipulators

1990 ◽  
Vol 112 (4) ◽  
pp. 653-660 ◽  
Author(s):  
H. Kazerooni ◽  
K. G. Bouklas ◽  
J. Guo
Keyword(s):  
Degrees Of Freedom ◽  
Control Method ◽  
Robot Manipulators ◽  
Industrial Robot ◽  
Redundant Robot ◽  
Control Approach ◽  
Six Degrees Of Freedom ◽  
The Stability ◽  
Control Methodology ◽  
Theory And Experiments

This work presents a control methodology for compliant motion in redundant robot manipulators. This control approach takes advantage of the redundancy in the robot’s degrees of freedom: while a maximum six degrees of freedom of the robot control the robot’s endpoint position, the remaining degrees of freedom impose an appropriate force on the environment. To verify the applicability of this control method, an active end-effector is mounted on an industrial robot to generate redundancy in the degrees of freedom. A set of experiments are described to demonstrate the use of this control method in constrained maneuvers. The stability of the robot and the environment is analyzed.

Design, testing and evaluation of an end-effector for self-relocation

Robotica ◽  
2015 ◽  
Vol 34 (12) ◽  
pp. 2689-2728 ◽  
Author(s):  
Feng Han ◽  
Kui Sun ◽  
Yu Liu ◽  
Hong Liu
Keyword(s):  
Degrees Of Freedom ◽  
Physical Simulation ◽  
Industrial Robot ◽  
Space Environment ◽  
Air Bearing ◽  
End Effector ◽  
Six Degrees Of Freedom ◽  
Soft Capture ◽  
End Effectors ◽  
Prototype Design

SUMMARYTwo identical end-effectors are indispensable for self-relocation of a space manipulator, which is an effective way of extending its servicing capability. The prototype design is intimately linked to the requirements. The significant features and functionality of the end-effector and its grapple fixture are described, including the key analysis efforts. The characteristics of the end-effector and their suitability for self-relocation and payload handling were confirmed by testing, which used two prototype end-effectors, a semi-physical simulation testbed system with two, six degrees of freedom (DOF) industrial robot arms, and an air-bearing testbed system with a seven DOF manipulator. The results demonstrate that the end-effector satisfies the requirements and it can work well in a simulated space environment. With the compliance motion of the manipulator, the end-effector can perform soft capture and the manipulator can securely self-relocate and handle the payload.

Sign in / Sign up

Export Citation Format

Share Document