1 of 34

Project 180: Smart Clothing for Kinematic and Dynamic Movements

Advisor: Professor Bunch

Client: Professor Tron

Gabrielle Bartolomei

Andrew Chen

Izabell Garcia

Wes Yan

Ilyoung Yang

May 2nd, 2025

2 of 34

Project 180

2

¼ Elderly individuals fall each year

3 million Emergency room visits

1 million hospitalizations

3 of 34

Project 180

The Problem

3

Previous Solutions: Computer Vision and EM systems → intrusive and require complex set up

The need: accurate, monitoring of mobility decline in elderly patients outside of a controlled environment

The Process:

Embed sensors

Data collection and storage

Evaluate accuracy of system

Integration of data to create postural pose

0.5 minutes

The problem that we are looking to address is that past solutions such as computer vision and electromagnetic systems, while providing accurate data, are intrusive and require complex set up, therefore cannot easily be translated to uncontrolled environments, such as the home

The need: accurate, noninvasive monitoring of mobility decline in elderly patients outside of a controlled environment

We broke this down into 4 steps including: embedding sensors, collecting and storing data, evaluating the accuracy of the system, and integrating the data to create a postural pose

1 minute: INTRO/BACKGROUND/PROBLEM STATEMENT

This project is a continuation and adaptation of Dr Roberto Tron’s past research investigating mobility monitoring of elderly aged patients → Izabel will further explain later in the presentation

Was proposed as a 4 part project:

Embedding of sensors within a system
Data collection and storage
Integration of data to create a postural pose
Evaluate the accuracy of the system

From these 4 objectives, there are 3 deliverables:

Determination of feasibility
Physical prototype
Build guide characterizing this device

Our group chose this project based on a variety of previous experiences including work with bendlab sensors in a laboratory setting, an internship with large scale data collection, and a shared interest in wearable health monitoring devices

-Motivation for the Project, why it matters, why you chose this project for your Capstone Project, history, where it fits into some larger context, why should anyone care.

4 of 34

Project 180

Customer Requirements and Specifications

4

Professor Tron’s Research:

Elderly stroke patients would have trouble moving their limbs. This trouble could lead to further injury and health complications in the future. Solved by using a deep learning camera and electrostimulation suits equided with accelerometers to help reduce muscle atrophy

Focuses to make system more user friendly and accessible:

Battery Life

Charging Time

Calibration

Durability

System Storage Capacity

Set up Time

Accuracy of Joint Angle Measurement Accuracy of Movement Speed

Pitfalls

Privacy/Invasive
Improper use of equipment
Accessibility
Environment Diversity

Benefits

Accurate Data
Instantaneous Response
Retain Muscle Memory
Personalization

5 of 34

Project 180

5

Conceptual Design Summary

Bendlab

IMU

ESP-32

BendLab:

4, outside of both knees and hips
Returns an angle

IMU BN055:

1, center of lower back
Returns angular velocity, acceleration, and orientation

ESP 32:

1, located in pocket on left upper thigh
Programmable light user interface

Battery Compartment

Esp32 and SD Card Compartment

Hole for User interface Light System

Housing:

For our conceptual design we researched past products and decided to incorporate 2 different sensors to achieve the desired metrics:

BendLab Capacitance sensor → precise joint angle measurement, while seamlessly integrating into the system due to their small profile, 4 in total, located on the outside of knees and hips

Critical for short term data, walking, movements, one side to other side comparison

IMU BN055 → returns angular velocity, acceleration, and orientation data, self calibrating

Can be used to verify bendlab data as well as long term orientation and postural states

Data will be processed via the ESP 32, a compact microcontroller (6mm x 6mm x 5 mm), with robust GPIO support, and wifi and bluetooth connectivity and data will be stored via a micro SD card. This is all stored in a 3D printed housing shown on the right. This enclosure additionally contains a hole that allows the light on the board to shine through and provide the user updates on operational status.

Alvin - 1 minute

*"On slide 14, we detail the primary components that make up our prototype. The sensing system includes BendLab capacitance sensors, a BNO055 IMU, and thin-film pressure sensors. Each was chosen for its specific strengths:

The BendLab sensors offer precise joint angle measurements while integrating seamlessly into the leggings.
The BNO055 IMU provides versatile posture and gait data with self-calibration features.
The pressure sensors enable accurate gait analysis with a small footprint that fits comfortably in the sole of the leggings.

These sensors work in tandem, providing robust data while maintaining wearer comfort."*

"For data processing, we selected the ESP-32 microcontroller due to its powerful yet compact design. It features built-in Wi-Fi and Bluetooth, expandable memory, and a low power profile—critical for continuous monitoring. This ensures data is not only stored efficiently but is ready for future real-time connectivity."

"On the user interface side, we paired voice guidance with LED indicators. This dual system ensures accessibility for users with varying needs while providing clear, real-time feedback."�Nice wes goodjob did chat write this for u?

6 of 34

Project 180

6

Conceptual Design Summary

Concept 1: Harness Concept 2: Leggings Concept 3: Compression Socks

7 of 34

Project 180

7

Prototype Generation:

Components:

4 BendLab
1 IMU
1 Control Housing

Fabrication:

Flat-felled seeming
Zigzag stitch
Removable backing

Alterations:

Velcro straps and pockets
Belt loops
Fabric Flap secured via velcro and snap buttons

8 of 34

Project 180

8

Prototype Generation

9 of 34

Project 180

9

Prototype Generation: Sensor Array

Final Sensor Implementation

Housing and User Interface

Final Prototype Assembly

10 of 34

Project 180

10

Prototype Evaluation: Ground Truth Testing

Testing Setup:

ACL Brace placed over Bendlab Sensor to compare joint angle measurements

11 of 34

Project 180

11

Prototype Evaluation: Functionality Tests

Sit to Stand Test

Sit Position

Knee Angle

~ 70°- 100°

Hip Angle

~ 35°

Stand Position

Knee Angle

~ 20°

Hip Angle

~ 20°

12 of 34

Project 180

12

Prototype Evaluation: Functionality Tests

Walking Test

Dynamic motion easily visualized

Knee and Hip angles coincide

Consistent angle readings in motion

13 of 34

Project 180

13

Prototype Evaluation: Prolonged Activity

IMU Orientation Angle Over Time

Flexion/Extension Angles

14 of 34

Project 180

14

Deliverables and Achievements

Long Battery Life

12 hours ~ 40%

Quick Calibration

~2 seconds

Easy to remove and charge battery

~5 seconds

Robust Data Storage

16 Gigabytes

Accurate Bend Angle Measurement

+/- 5 degrees

Minimal System Drift

BendLab: <1 degree

IMU: ~10 degrees

15 of 34

Project 180

15

Takeaways and Recommendations

01

Improve mechanical protection of BendLab sensors and connection robustness

02

Strengthen waterproofing through sealed housing, embedded electronics, or coatings to allow for pre-positioned sensors

03

Pre-positioned sensors would decrease user error, increase ease of use, and increase accuracy of measurements

BendLab sensors are highly sensitive

Long setup time and specific sensor placement

Modular design allows for greater precision than accuracy

16 of 34

Questions?

Thank You

Professor Bunch

Professor Tron

Nick Rudh

BU Soft Robotics Control Lab

Project 180

16

17 of 34

Citations

[1] McArdle, W.D.; Katch, F.I.; Katch, V.L. “Essentials of Exercise Physiology;” Lippincott Williams & Wilkins: Philadelphia, PA, USA, 2006.

[2] Edriss, S.; Romagnoli, C.; Caprioli, L.; Zanela, A.; Panichi, E.; Campoli, F.; Padua, E.; Annino, G.; Bonaiuto, V. “The Role of Emergent Technologies in the Dynamic and Kinematic Assessment of Human Movement in Sport and Clinical Applications.” Appl. Sci. 2024, 14, 1012. https://doi.org/10.3390/app14031012

[3] Tan, Emily. “Roberto Tron: Redefining Health Care for Aging America.” College of Engineering Roberto Tron Redefining Health Care for Aging America Comments, Sept. 2024.

[4] Jones CH, Dolsten M. Healthcare on the brink: navigating the challenges of an aging society in the United States. NPJ Aging. 2024 Apr 6;10(1):22. doi: 10.1038/s41514-024-00148-2. Erratum in: NPJ Aging. 2024 May 10;10(1):25. doi: 10.1038/s41514-024-00153-5. PMID: 38582901; PMCID: PMC10998868.

[5] Tron, Roberto. Customer Interview. Interview by G. Bartolomei, A. Chen, I. Garcia, W. Yan, I. Yang, 24 Sept. 2024. Personal interview.

[6] Mesquita, Inês Albuquerque, Pedro Filipe Pereira da Fonseca, Ana Rita Vieira Pinheiro, Miguel Fernando Paiva Velhote Correia, and Cláudia Isabel Costa da Silva. 2019. “Methodological Considerations for Kinematic Analysis of Upper Limbs in Healthy and Poststroke Adults Part II: A Systematic Review of Motion Capture Systems and Kinematic Metrics.” Topics in Stroke Rehabilitation 26 (6): 464–72. doi:10.1080/10749357.2019.1611221.

[7] Samatas, G.G.; Pachidis, T.P. “Inertial Measurement Units (IMUs) in Mobile Robots over the Last Five Years: A Review”. Designs 2022, 6, 17. doi.org/10.3390/designs6010017

[8] A. M. Franz, T. Haidegger, W. Birkfellner, K. Cleary, T. M. Peters and L. Maier-Hein, "Electromagnetic Tracking in Medicine—A Review of Technology, Validation, and Applications," in IEEE Transactions on Medical Imaging, vol. 33, no. 8, pp. 1702-1725, Aug. 2014, doi: 10.1109/TMI.2014.2321777.

[9] Robert A. Conner, (2019). Smart clothing for ambulatory motion capture (US10321873B2)

[10] Robert A. Conner, (2020). Smart clothing with converging/diverging bend or stretch sensors for measuring body motion or configuration (US10716510B2)

[11] Geof Auchinlek, Paul Sharman (2017). Method for calibrating apparatus or monitoring rehabilitation from joint surgery. (US10456075B2)

[12] Bendlabs, “Soft Angular Displacement Sensor Theory Manual,” ad Theory Guide, June 2022

[13] Bendlabs, “2-Axis Datasheet,” Bend Labs Flexible Soft Sensor (2-axis), June 2022

[14] Jing, Ran. BendLab Sensor Interview. Interviewed by Ilyoung Yang. 12 Oct. 2024. Personal Interview

18 of 34