+1443 776-2705 panelessays@gmail.com
  

I have some comments for things I would like to be changed in/added to my presentation:

1)  Add real picture of a robot in slide 1 & 2

2)  Slide 5 should have the title (Problem Statement), and Slide 6&7 are just continuation

3)  Specify that the used method will be Optimization

4)  Missing Point should be included that is the Input Data 

5)  In Slide 8 , the last point is good but the others needs review 

6)  In Slide 10 change the title to (Literature Review), This Part should be longer with more info

7)  In Slide 11, I need some questions to make the methodology clear

8)  In Slide 12, some results(Graphs) should be added from other authors and analyse them

9)  If there was any extra references used then it should be added to the ppt

Lastly I provided examples along with (( My Presentation(Luna) ))

Safe robot operation using force tracking impedance control

Luna

1

contents

Introduction

Background

Impedance Control

Problem Statement

Challenges

Research Question

Aim

Objectives

Significance of the study

2

Expected Outcomes

Literature Review

Methodology

Expected Results and Discussion

Conclusion

Introduction

The scope of industrial robot applications was established from the conventional handling, assembly and welding tasks leading to a wide range of production.

Such applications involves robot end-effector interaction with environment.

In interaction case, insufficient compliance for manipulator is always a key problem.

Only position control is not sufficient to control and handle the interaction.

It is necessary to develop an interaction control method that achieves position tracking and reliably adapts the force exerted (force tracking) on the environment in order to avoid damage to both environment and the manipulator.

Impedance control is a feasible solution to overcome position uncertainties while robot-environment interaction and avoid large impact forces

3

3

Impedance control

Impedance controller resembles a virtual mass-spring-damper system between the environment and robot end-effector

This allows the robot to safely interacts with the environment.

Mechanical impedance is the ratio of force input to position output.

Impedance shows that how much the body resist the applied forces.

Mass-spring-damper system while interaction

4

Challenges in impedance control

Impedance model is second order equation with desired impedance parameters.

Require optimal tuned parameters to achieve the desired interactive impedance.

To avoid parameters tuning, additional control algorithm is implemented.

This algorithm helps in minimizing the force tracking error.

5

Problem Statement

Research Question

How the impedance control can be improved to enhance robot operation safety while environment interaction.

Research Aim

The aim of this research is to

Improve the force tracking impedance control structure for enhanced end-effector force tracking.

Furthermore, optimally tune the impedance parameter to make the system more sensitive for small force change.

6

Problem Statement

Research Objective

To perform the proposed research, the research objectives are

Review the literature to understand the impedance model

Then update the impedance model in a way that will improve the end-effector force tracking.

Implement machine learning or optimization algorithm to optimally tune the impedance parameters.

Implement and compare the proposed and existing algorithms.

7

Significance of study

For robust robot operation while interaction, integrated position and force control is utilized.

Usually, the position control is focused for improved performance. Whereas for force control, traditional impedance model is used.

Therefore, sometimes impedance results are not precise in the presence of external disturbance.

This research will then contribute specially to impedance control.

This will help in enhance the robot safety specially when human and robot are working together.

8

Expected outcomes

Improvement and Implementation of Impedance Control

Will enhance the robot end-effector trajectory tracking while interaction.

This will improve the safety of robot operation

Resulting in avoiding damage to both the robot and the environment.

9

Optimal impedance parameters

Will make system more sensitive to small end-effector forces.

This will help the system to easily work in sensitive environments.

Impedance Control literature

Year Author Contribution
1985 Hogan Hybrid position/force control
1985 Hogan Impedance control
1991 Chan et al. SMC-based impedance control
1997 Seraji et al. Force tracking in impedance control
2003 Iwasaki et al. Adaptive force control using sliding mode control
2004 Jung et al. Adaptive impedance force tracking under unknown environment
2008 Lee et al. Force tracking with variable target stiffness
2018 Liang et al. Force tracking with unknown environment via iterative learning algorithm
2019 Yang et al. PID-Based force tracking with nonlinear velocity observer

10

References

N. Hogan, “Impedance control: An approach to manipulation: Part I—Theory,” 1985.

N. Hogan, “Impedance control: An approach to manipulation: Part II—Implementation,” 1985.

F. J. Abu-Dakka, and M. Saveriano, “Variable impedance control and learning – a review” frontiers in Robotics and AI, vol. 7, 590681, 2020

S. Chan, B. Yao, W. Gao et al., “Robust impedance control of robot manipulators,” Robots Automation, vol. 6, no. 4, pp. 220–227, 1991.

J. Peng, Z. Yang, and T. Ma, “Position/force tracking impedance control for robot systems with uncertainties based on adaptive jacobian and neural network,” complexity, vol. 2019, p. 1406534.

11

methodology

The methodology involves the following steps

Impedance Model Improvement: Analysis of different control algorithm and their comparison will help in the selection of best algorithm to improve the impedance model.

Optimization of Impedance Parameters: Utilize the machine learning or optimization algorithm such as Particle Swarm Optimization (PSO) for optimal impedance parameters.

Implementation and Comparison: Implement the final algorithms and compare with existing algorithms.

12

Expected Results and discussion

After the research, the results and discussions are expected as

The improved impedance control enhanced the robot end-effector force tracking by converging the force tracking error to zero.

This is because the control algorithm integrated in impedance control provides more energy to the robot.

Furthermore, the optimization technique helped the robot detect the small change in the end-effector force based on changing environment.

This is because the algorithm has optimally tuned the parameters based on the environment dynamics.

13

conclusion

Insufficient compliance while robot-environment contact is always a key problem.

This research presented a basic and easy review about the impedance control.

Mainly, conventional impedance control with adaptive algorithm and impedance control with modified position control loop is utilized.

Improved force control loop was very useful in making the robot operation safer as required.

Optimized desired impedance parameters avoided sudden force impact by detecting small force change.

14

References

N. Hogan, “Impedance control: An approach to manipulation: Part I—Theory,” 1985.

N. Hogan, “Impedance control: An approach to manipulation: Part II—Implementation,” 1985.

S. Chan, B. Yao, W. Gao et al., “Robust impedance control of robot manipulators,” Robots Automation, vol. 6, no. 4, pp. 220–227, 1991.

H. Seraji and R. Colbaugh, “Force tracking in impedance control,” The International Journal of Robots Research, vol. 16, no. 1, pp. 97-117, 1997.

M. Iwasaki, N. Tsujiuchi, and T. Koizumi, “Adaptive force control for unknown environment using sliding mode control with variable hyperplane,” JSME International Journal Series C Mechanical Systems, Machine Elements and Manufacturing, vol. 46, no. 3, pp. 967-972, 2003.

S. Jung, T. C. Hsia, and R. G. Bonitz, “Force tracking impedance control of robot manipulators under unknown environment,” IEEE Transactions on Control Systems Technology, vol. 12, no. 3, pp. 474-483, 2004.

X. Liang, H. Zhao, X. Li, and H. Ding, “Force tracking impedance control with unknown environment via an iterative learning algorithm,” in 2018 3rd International Conference on Advanced Robots and Mechatronics (ICARM), 2018: IEEE, pp. 158-164.

J. Peng, Z. Yang, and T. Ma, “Position/force tracking impedance control for robot systems with uncertainties based on adaptive jacobian and neural network,” complexity, vol. 2019, p. 1406534.

15

Thank You

16

Prediction of Early Childhood Obesity in Saudi Arabia using Machine Learning

1

Outline

Significance of the Study

03

Expected Outcomes

04

01

Introduction and Background

02

Problem Statement

Literature Review

05

Draft Conclusion

08

2

Methodology

06

Expected Results and Discussion

07

Research Question, Aim, Objectives

March 4, 2021 CS3172

3

Introduction and Background

Obesity

Effects

Type 2 Diabetes

Heart

Diseases

Cancers

Strokes

2.1 billion people are obese.

42 million children under the age of 5 were overweight in 2013.

30% of the global population.

41% increase by 2030 if current trend persists.

Risk Factors:

Unhealthy eating patterns

Lack of exercise

Genetics

Psychological factors

Socioeconomic factors

Obesity, a global epidemic.

March 4, 2021 CS3172

M. H. M. Adnan, W. Husain, & Rashid, N. A. (). A framework for childhood obesity classifications and predictions using NBtree. 2011 7th International Conference on Information Technology in Asia, (), 1–6. https://doi.org/10.1109/CITA.2011.5999502

Child Health – Childhood Obesity. (2019, November 21). Www.moh.gov.sa. https://www.moh.gov.sa/en/HealthAwareness/EducationalContent/BabyHealth/Pages/003.aspx

Al-Hussaini, A., Bashir, M., Khormi, M., AlTuraiki, M., Alkhamis, W., Alrajhi, M., & Halal, T. (2019). Overweight and obesity among Saudi children and adolescents: Where do we stand today? Saudi Journal of Gastroenterology, 25(4), 229. https://doi.org/10.4103/sjg.sjg_617_18

Obesity In Saudi Arabia

4

Introduction and Background

Pediatric obesity

Adult obesity

Prevention of childhood obesity is urgently required for reduction in obesity prevalence.

Childhood obesity prevalence was 18.2% in 2019.

March 4, 2021 CS3172

Obesity – Adults (18+ years) | EMRO Regional Health Observatory. (2017). Rho.emro.who.int. https://rho.emro.who.int/ThemeViz/TermID/146

Graph from:

GHO | By category | Prevalence of obesity among adults, BMI = 30, age-standardized – Estimates by country. (2017). WHO. https://apps.who.int/gho/data/node.main.A900A?lang=en

Al-Hussaini, A., Bashir, M., Khormi, M., AlTuraiki, M., Alkhamis, W., Alrajhi, M., & Halal, T. (2019). Overweight and obesity among Saudi children and adolescents: Where do we stand today? Saudi Journal of Gastroenterology, 25(4), 229. https://doi.org/10.4103/sjg.sjg_617_18

^ for childhood rate

5

Introduction and Background

Early Childhood Obesity Prediction System

Diagnosis

Prediction

Classification

Indication

Genetics

BMI

ML

How can we predict?

Machine learning is the study of computer algorithms that improve automatically through experience. It is seen as a part of artificial intelligence.

March 4, 2021 CS3172

High Body Mass Index in Infancy May Predict Severe Obesity in Early Childhood

The prevalence of overweight and obesity has risen alarmingly among Saudi children and adolescents over the past decade and should make a strong case to initiate and monitor effective implementation of obesity prevention measures.

2. Problem Statement

How can Machine Learning be used to predict early childhood obesity in

Saudi Arabia?

6

2.1 Research Question

March 4, 2021 CS3172

2. Problem Statement

7

Construct a Machine Learning model to predict the risk of early childhood obesity in Saudi Arabia.

2.2 Research Aim

March 4, 2021 CS3172

Aim: The aim of the study is to predict early childhood obesity and the reduce it risk.

2. Problem Statement

8

Implement a machine learning algorithm suitable for prediction.

Collect and Analyze the clinical data including weight, height, BMI, socioeconomic conditions.

Compare and contrast developed prediction model with existing ones.

2.3 Research Objectives

Develop a prediction model for early children obesity.

March 4, 2021 CS3172

Three simple factors can predict whether a child is likely to be overweight or obese by the time they reach adolescence: the child’s body mass index (BMI), the mother’s BMI and the mother’s education level, according to our new research.

Predicting obesity in children using clinical data before the age of two using machine learning techniques.

Investigate the ability of the child’s BMI and the mother’s BMI to predict. To relate the calculated BMI of the child and mother.

Advance our understanding towards the development of smart and effective interventions for childhood obesity care. (from another paper)

Develop a prediction model to help clinicians identify candidate children for early obesity interventions, thereby targeting at risk children at a critical age of development related to establishing eating and lifestyle habits.

About data?

Identify

evaluate

3. Significance of the Study

Contribution to vision 2030 goal of Focus on Wellbeing and Preventive Care.

Contribution to SDG 17 goal of Good Health and Well-being.

Helps in early medical intervention to prevent obesity in early stages leading to a healthier society.

9

Client

March 4, 2021 CS3172

Vision 2030: Focus on Wellness and Preventive Care: The Saudi Government has rolled out initiatives focusing on fitness and preventive care. KSA is aiming for a 3 per cent reduction in obesity by 2030.

https://www.globalhealthsaudi.com/en/overview/saudi-news/Top-priorities-in-KSAs-healthcare-sector.html

Contribution to vision 2030 goal of Focus on Wellbeing and Preventive Care.

Contribution to SDG 17 goal of Good Health and Well-being.

Targeting a large group of children suffering from the most widely spread illnesses in the world,

Children who have obesity are more likely to have: High blood pressure and high cholesterol, which are risk factors for cardiovascular disease.Childhood obesity is associated with a higher chance of premature death and disability in adulthood.

Dont let the adults to reached this sitution.

The numbers are increasing

Preventing is better then treatment.

Focusing on these causes may, over time, decrease childhood obesity and lead to a healthier society as a whole.

For healthcare people

4. Expected Outcomes

10

Development of a computer code for early childhood obesity prediction using machine learning techniques

Identification of potentially high-risk children to whom future obesity prevention strategies should be applied

March 4, 2021 CS3172

2. For the clients which are doctors so that they can stop

5. Literature Review

11

Machine Learning

Predicting obesity is a difficult task and machine learning techniques are powerful for this task.

Prediction Models

Data, BMI and genetics, are good predictors and childhood obesity models can be valuable assets to healthcare applications.

Increasing Obesity Rates

Rates of obesity among Saudi children are significantly increasing.

Current models have low accuracy

Several approaches to early childhood obesity are proposed, but they are inaccurate and no research targets Saudi Arabia.

March 4, 2021 CS3172

Several models have been developed

Current models are inaccurate

No prediction models targeting Saudi Arabia’s children.

Only 2 models are currently in use in the US.

Several approaches to early childhood obesity are proposed, but they are inaccurate.

Prediction models can be deduced from risk factors and genetics.

Childhood obesity models can be valuable assets to healthcare applications.

5. Literature Review

12

Research is limited.

Poor prediction rates.

Early-stage point.

Zhang, S., Tjortjis, C., Zeng, X. et al.:

Evaluated: well-known data mining algorithms

Result: SVM and Bayesian algorithms are the best algorithms for predicting obesity.

Universiti Sains Malaysia:

Proposed: hybrid approach using Naïve Bayes and Genetic Algorithm

Result: 75% accuracy improvement over Naive Bayes.

Indiana University:

Test: six ML algorithms

Result: NaiveBayes and BayesNet high accuracy at 85% and sensitivity at 90% respectively.

Universiti Sains Malaysia:

Data mining techniques on data of ages 9 to 11

Proposed: more significant parameters on an existing solution

Result: 21% accuracy improvement.

Gaps

March 4, 2021 CS3172

That after adding more parameters to an existing solution, the algorithm’s result predictions of childhood obesity were 21% more accurate than before adding them

Research is limited: and data used are of ages where obesity is well established and harder to remediate.

Poor prediction rates that differ from test to test.

Early stage point : Research is at an early point as they only recently started predicting obesity using ML where more works have to be done in the future.

6. Methodology

Utilize the amount of clinical data including BMI, family history, and lifestyle habits.

Data

Prediction Model

13

Analysis of different machine learning methods such as RandomTree, RandomForest, Naïve Bayes, and others trained on collected data.

Develop an algorithm to accurately predict obesity of children based on Naive Bayes using Python.

Machine Learning

(Singh & Tawfik, 2020)

March 4, 2021 CS3172

Database and Machine learning

Program???

Singh, B., & Tawfik, H. (2020). Machine Learning Approach for the Early Prediction of the Risk of Overweight and Obesity in Young People. Lecture Notes in Computer Science, 523–535. https://doi.org/10.1007/978-3-030-50423-6_39

Naive Bayes is a simple but surprisingly powerful algorithm for predictive modeling and based on the literature review, Naive Bayes was food to have the highest levels of accuracy among others.

Some intervention measures such as physical activity and healthy eating can be a fundamental component to maintain a healthy lifestyle. Therefore, it is absolutely essential to detect childhood obesity as early as possible.

https://data.humdata.org/dataset/who-data-for-saudi-arabia

https://childmortality.org/data/Saudi%20Arabia

https://data.unicef.org/country/sau/

https://data.worldbank.org/indicator/SH.STA.OWGH.ZS

https://data.gov.sa/Data/en/dataset

7. Expected Results and Discussion

14

Submission of a coded program runned using a machine learning algorithm that takes as input children’s data before two years old and predict if they are at risk of developing obesity when they get older.

A comparison of accuracy results between proposed developed solution with existing ones that use different machine learning algorithms.

Implementation of the developed program in hospitals to help physicians predict future occurrences of obesity in children.

March 4, 2021 CS3172

Such machine learning models may be valuable as part of a healthcare application that alerts individuals with high risk of obesity incidence in the future.

8. Draft Conclusion

Predicting obesity at an early age is both useful and important because preventive

measures and proper interventions can be applied if the children indicated a high risk of obesity.

The proposed early childhood obesity prediction model:

Implement an ML algorithm called Naive Bayes.

Predict the risk of obesity in early childhood.

15

March 4, 2021 CS3172

Limitations:

Difficult to get data from Saudi Arabia (trying to find global database and maybe late try to localize it)

Thank You

Any Questions?

16

References

Al-Hussaini, A., Bashir, M., Khormi, M., AlTuraiki, M., Alkhamis, W., Alrajhi, M., & Halal, T. (2019). Overweight and obesity among Saudi children and adolescents: Where do we stand today? Saudi Journal of Gastroenterology, 25(4), 229. https://doi.org/10.4103/sjg.sjg_617_18

CDC. (2019). Defining Childhood Obesity. Centers for Disease Control and Prevention. https://www.cdc.gov/obesity/childhood/defining.html

Child Health – Childhood Obesity. (2019, November 21). Www.moh.gov.sa. https://www.moh.gov.sa/en/HealthAwareness/EducationalContent/BabyHealth/Pages/003.aspx

GHO | By category | Prevalence of obesity among adults, BMI = 30, age-standardized – Estimates by country. (2017). WHO. https://apps.who.int/gho/data/node.main.A900A?lang=en

M. H. M. Adnan, W. Husain, & Rashid, N. A. (). A framework for childhood obesity classifications and predictions using NBtree. 2011 7th International Conference on Information Technology in Asia, (), 1–6. https://doi.org/10.1109/CITA.2011.5999502

Obesity – Adults (18+ years) | EMRO Regional Health Observatory. (2017). Rho.emro.who.int. https://rho.emro.who.int/ThemeViz/TermID/146

Saad, A. (2018). Prevention of Childhood Obesity in Saudi Arabia. J Child Obes, 2-002. https://doi.org/10.21767/2572-5394.100057

Singh, B., & Tawfik, H. (2020). Machine Learning Approach for the Early Prediction of the Risk of Overweight and Obesity in Young People. Lecture Notes in Computer Science, 523–535. https://doi.org/10.1007/978-3-030-50423-6_39

Smego, A., Woo, J. G., Klein, J., Suh, C., Danesh Bansal, Bliss, S., Daniels, S. R., Bolling, C., & Crimmins, N. A. (2017). High body mass index in infancy may predict severe obesity in early childhood. The Journal of Pediatrics, 183, 87-93.e1. https://doi.org/https://doi.org/10.1016/j.jpeds.2016.11.020

Presenting a prototype of Li-Fi for smart cities

Why Li-Fi and how to improve it ?

Outline

2

Introduction & Background

Problem Statement

Research Question & aim

Significance

Contribution

Methodology

Expected Results

Literature Review

Limitations

Timeline

Results

Conclusion

Q&A

References

Background

Connectivity

Current need of h