Optimization of Grasp Points in Robotic Manipulation
Grasp Points Optimization in Multi-
fingered Grasping
聂亦丁
: 12111826 |
刘冠骅
: 12010711
孔志伟
: 12111205 | 范彧恒: 12111034
王俊麟
: 12112921
|
黄硕
: 12111608
Group# 3
2024.3.26
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Proposed Project Title Summary
•
Summarize the whole summary in a single paragraph here, within 300~400
words, describing the following contents.
•
(20%) What is the problem that you will be investigating? Why is it interesting?
•
(20%) What reading will you examine to provide context and background?
•
(20%) What data will you use? If you are collecting new data, how will you do it?
•
(20%) What method or algorithm are you proposing? If there are existing
implementations, will you use them, and how? How do you plan to improve or modify
such implementations? You don't have to have an exact answer at this point, but you
should have a general sense of how you will approach the problem you are working on.
•
(20%) How will you evaluate your results? Qualitatively, what kind of results do you
expect (e.g., plots or figures)? Quantitatively, what kind of analysis will you use to
evaluate and/or compare your results (e.g., what performance metrics or statistical tests)?
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
2
Proposed Project Title Summary
Grasp Points Optimization in Multi-fingered Grasping
Grasp points optimization is to solve the optimal positions where forces of the gripper are applied. This optimization offers initial target
points to the robot, making further manipulations possible. For most objects, they are irregular and only provide few suitable position to fetch.
Consequently, it is necessary and interesting to find a solution to optimize the grasp-point by minimizing forces and torques while maximizing
stability for robotics.
We shall draw upon the approach delineated in the paper titled Computation of multi-fingered grasping force with linear combination and
the concept presented in the paper titled Robotic Grasp Detection using Deep Convolutional Neural Networks and Multi-view Self-supervised
Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge.
We plan to build a simulation environment. In this environment, we will pre configure some representative objects for the robotic arm to
grip, such as cones, cylinders, etc.. In this process, we will use the known surface and position information of the object as data input, and we will
determine whether our algorithm is reasonable by observing the gripping performance of the robotic arm.
Computation of multi-fingered grasping force with linear combination
by CHEN Dongjin，JIANG Li and WANG Xinqing provides an
efficient method to solve the optimal force distribution of grasp positions, which is usually a non-statically
determined problem, using linear combination of unit external force and their unit response. Through this method, we
can generate initial values for the force optimization algorithms in the point contact friction models, and improve
efficiency.
Furthermore, we plan to use machine learning to help us find the best grasp-point for objects with clear known
shape and position information in a simulated environment, where the initial position parameters of machine learning
are the initial values obtained above. The methodology can be referred from the paper
Robotic Grasp Detection using Deep
Convolutional Neural Networks.
Quantitatively, as previously indicated, the effectiveness of our algorithm will be evaluated through the observation and analysis of the
robotic arm's grasping performance. Qualitatively, we will assess the stability and time efficiency throughout the gripping process executed by the
robotic arm. We intend to compile tables that succinctly delineate these performance outcomes. Additionally, we may supplement our findings
with accompanying videos to substantiate our results.
We will start from 2D-2 points grasping, and we expect to offer a functioning AI to solve a 3D-N points grasping problem.
3
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
What is the problem that you will be investigating?
Grasp points optimization is to solve the optimal positions where forces of the
gripper are applied. This optimization offers initial target points to the robot, making
further manipulations possible. For most objects, they are irregular and only provide
few suitable position to fetch. Consequently, it is necessary and interesting to find a
solution to optimize the grasp-point by minimizing forces and torques while
maximizing stability for robotics.
Why is it interesting?
4
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
What reading will you examine?
We shall draw upon the approach delineated in the paper titled
Computation of multi-
fingered grasping force with linear combination
and the concept presented in the paper titled
Robotic Grasp Detection using Deep Convolutional Neural Networks
and
Multi-view Self-
supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge
.
To provide context and background
5
Multi-view Self-supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge
：
Zeng, A., Yu, K.T., Song, S., Suo, D., Walker, E., Rodriguez, A. and Xiao, J., 2017, May. Multi-view self-supervised deep learning for 6d pose
estimation in the amazon picking challenge. In 2017 IEEE international conference on robotics and automation (ICRA) (pp. 1386-1383). IEEE.
Computation of multi-fingered grasping force with linear combination
：
https://max.book118.com/html/2019/0115/6101222011002002.shtm
Robotic Grasp Detection using Deep Convolutional Neural Networks
S. Kumra and C. Kanan, "Robotic grasp detection using deep convolutional neural networks," 2017 IEEE/RSJ International Conference on Intelligent
Robots and Systems (IROS), Vancouver, BC, Canada, 2017, pp. 769-776, doi: 10.1109/IROS.2017.8202237. keywords: {Feature extraction;Robotkinematics;Machine learning;Grippers;Training}
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
What data will you use?
We plan to build a simulation environment. In this environment, we will pre configure
some representative objects for the robotic arm to grip, such as cones, cylinders, etc.. In
this process, we will use the known surface and position information of the object as data
input, and we will determine whether our algorithm is reasonable by observing the
gripping performance of the robotic arm.
If you are collecting new data, how will you do it?
6
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
What method or algorithm are you proposing?
Computation of multi-fingered grasping force with linear combination
by CHEN
Dongjin, JIANG Li and WANG Xinqing provides an efficient method to solve the
optimal force distribution of grasp positions, which is usually a non-statically
determined problem, using linear combination of unit external force and their unit
response. Through this method, we can generate initial values for the force optimization
algorithms in the point contact friction models, and improve efficiency.
Furthermore, we plan to use machine learning to help us find the best grasp-point
for objects with clear known shape and position information in a simulated
environment, where the initial position parameters of machine learning are the initial
values obtained above. The methodology can be referred from the paper
Robotic Grasp
Detection using Deep Convolutional Neural Networks.
If there are existing implementations, will you use them, and how? How do you plan to improve or modify such implementations? You don't have to
have an exact answer at this point, but you should have a general sense of how you will approach the problem you are working on.
7
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
How will you evaluate your results?
Quantitatively, as previously indicated, the effectiveness of our algorithm will be
evaluated through the observation and analysis of the robotic arm's grasping performance.
Qualitatively, we will assess the stability and time efficiency throughout the gripping
process executed by the robotic arm. We intend to compile tables that succinctly delineate
these performance outcomes. Additionally, we may supplement our findings with
accompanying videos to substantiate our results.
We will start from 2D-2 points grasping, and we expect to offer a functioning AI to
solve a 3D-N points grasping problem.
Qualitatively, what kind of results do you expect (e.g., plots or figures)? Quantitatively, what kind of analysis will you use to evaluate and/or compare
your results (e.g., what performance metrics or statistical tests)?
8
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
SUSTech Design + Learning Lab
PowerPoint Template
孔志伟
: 12111205
SUSTech
Chaoyang Song
12111205@mail.sustech.edu.cn
聂亦丁
: 12111826 |
刘冠骅
: 12010711
Grasp Points Optimization in Multi-fingered Grasping
Grasp points optimization in multi-fingered robotic grasping aims to find optimal positions for applying gripper forces, enabling efficient robotic manipulation of irregular objects while maximizing stability. Leveraging techniques from papers on multi-fingered force computation and robotic grasp detection using deep learning, a simulation environment will be created to evaluate grasp performance on representative objects. Machine learning will assist in determining grasp points for objects with known shapes and positions, improving efficiency and effectiveness through quantitative and qualitative assessments.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Grasp Points Optimization in Multi- fingered Grasping : 12111826 | : 12010711 : 12111205 | : 12111034 : 12112921 | : 12111608 Group# 3 2024.3.26 AncoraSIR.com : 12111826 | : 12010711
Proposed Project Title Summary Summarize the whole summary in a single paragraph here, within 300~400 words, describing the following contents. (20%) What is the problem that you will be investigating? Why is it interesting? (20%) What reading will you examine to provide context and background? (20%) What data will you use? If you are collecting new data, how will you do it? (20%) What method or algorithm are you proposing? If there are existing implementations, will you use them, and how? How do you plan to improve or modify such implementations? You don't have to have an exact answer at this point, but you should have a general sense of how you will approach the problem you are working on. (20%) How will you evaluate your results? Qualitatively, what kind of results do you expect (e.g., plots or figures)? Quantitatively, what kind of analysis will you use to evaluate and/or compare your results (e.g., what performance metrics or statistical tests)? AncoraSIR.com 2 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
Proposed Project Title Summary Grasp Points Optimization in Multi-fingered Grasping Grasp points optimization is to solve the optimal positions where forces of the gripper are applied. This optimization offers initial target points to the robot, making further manipulations possible. For most objects, they are irregular and only provide few suitable position to fetch. Consequently, it is necessary and interesting to find a solution to optimize the grasp-point by minimizing forces and torques while maximizing stability for robotics. We shall draw upon the approach delineated in the paper titled Computation of multi-fingered grasping force with linear combination and the concept presented in the paper titled Robotic Grasp Detection using Deep Convolutional Neural Networks and Multi-view Self-supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge. We plan to build a simulation environment. In this environment, we will pre configure some representative objects for the robotic arm to grip, such as cones, cylinders, etc.. In this process, we will use the known surface and position information of the object as data input, and we will determine whether our algorithm is reasonable by observing the gripping performance of the robotic arm. Computation of multi-fingered grasping force with linear combination by CHEN Dongjin JIANG Li and WANG Xinqing provides an efficient method to solve the optimal force distribution of grasp positions, which is usually a non-statically determined problem, using linear combination of unit external force and their unit response. Through this method, we can generate initial values for the force optimization algorithms in the point contact friction models, and improve efficiency. Furthermore, we plan to use machine learning to help us find the best grasp-point for objects with clear known shape and position information in a simulated environment, where the initial position parameters of machine learning are the initial values obtained above. The methodology can be referred from the paper Robotic Grasp Detection using Deep Convolutional Neural Networks. Quantitatively, as previously indicated, the effectiveness of our algorithm will be evaluated through the observation and analysis of the robotic arm's grasping performance. Qualitatively, we will assess the stability and time efficiency throughout the gripping process executed by the robotic arm. We intend to compile tables that succinctly delineate these performance outcomes. Additionally, we may supplement our findings with accompanying videos to substantiate our results. We will start from 2D-2 points grasping, and we expect to offer a functioning AI to solve a 3D-N points grasping problem. AncoraSIR.com 3 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
What is the problem that you will be investigating? Why is it interesting? Grasp points optimization is to solve the optimal positions where forces of the gripper are applied. This optimization offers initial target points to the robot, making further manipulations possible. For most objects, they are irregular and only provide few suitable position to fetch. Consequently, it is necessary and interesting to find a solution to optimize the grasp-point by minimizing forces and torques while maximizing stability for robotics. AncoraSIR.com 4 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
What reading will you examine? To provide context and background We shall draw upon the approach delineated in the paper titled Computation of multi- fingered grasping force with linear combination and the concept presented in the paper titled Robotic Grasp Detection using Deep Convolutional Neural Networks and Multi-view Self- supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge. Computation of multi-fingered grasping force with linear combination https://max.book118.com/html/2019/0115/6101222011002002.shtm Multi-view Self-supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge Zeng, A., Yu, K.T., Song, S., Suo, D., Walker, E., Rodriguez, A. and Xiao, J., 2017, May. Multi-view self-supervised deep learning for 6d pose estimation in the amazon picking challenge. In 2017 IEEE international conference on robotics and automation (ICRA) (pp. 1386-1383). IEEE. Robotic Grasp Detection using Deep Convolutional Neural Networks S. Kumra and C. Kanan, "Robotic grasp detection using deep convolutional neural networks," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, Canada, 2017, pp. 769-776, doi: 10.1109/IROS.2017.8202237. keywords: {Feature extraction;Robot kinematics;Machine learning;Grippers;Training} AncoraSIR.com 5 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
What data will you use? If you are collecting new data, how will you do it? We plan to build a simulation environment. In this environment, we will pre configure some representative objects for the robotic arm to grip, such as cones, cylinders, etc.. In this process, we will use the known surface and position information of the object as data input, and we will determine whether our algorithm is reasonable by observing the gripping performance of the robotic arm. AncoraSIR.com 6 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
What method or algorithm are you proposing? If there are existing implementations, will you use them, and how? How do you plan to improve or modify such implementations? You don't have to have an exact answer at this point, but you should have a general sense of how you will approach the problem you are working on. Computation of multi-fingered grasping force with linear combination by CHEN Dongjin, JIANG Li and WANG Xinqing provides an efficient method to solve the optimal force distribution of grasp positions, which is usually a non-statically determined problem, using linear combination of unit external force and their unit response. Through this method, we can generate initial values for the force optimization algorithms in the point contact friction models, and improve efficiency. Furthermore, we plan to use machine learning to help us find the best grasp-point for objects with clear known shape and position information in a simulated environment, where the initial position parameters of machine learning are the initial values obtained above. The methodology can be referred from the paper Robotic Grasp Detection using Deep Convolutional Neural Networks. AncoraSIR.com 7 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
How will you evaluate your results? Qualitatively, what kind of results do you expect (e.g., plots or figures)? Quantitatively, what kind of analysis will you use to evaluate and/or compare your results (e.g., what performance metrics or statistical tests)? Quantitatively, as previously indicated, the effectiveness of our algorithm will be evaluated through the observation and analysis of the robotic arm's grasping performance. Qualitatively, we will assess the stability and time efficiency throughout the gripping process executed by the robotic arm. We intend to compile tables that succinctly delineate these performance outcomes. Additionally, we may supplement our findings with accompanying videos to substantiate our results. We will start from 2D-2 points grasping, and we expect to offer a functioning AI to solve a 3D-N points grasping problem. AncoraSIR.com 8 Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
SUSTech Design + Learning Lab PowerPoint Template : 12111205 SUSTech Chaoyang Song 12111205@mail.sustech.edu.cn AncoraSIR.com Grasp Points Optimization in Multi-fingered Grasping : 12111826 | : 12010711
