OpenExo GUI¶

Overview¶

This application runs on any system with Python 3.12 installed. Required libraries can be installed via pip.

Open Source Application utilizing the OpenExo API to search, connect to, and control NAU Biomechatronics Lab’s OpenExo systems.

This project is licensed under the MIT License and provides an open-source solution for accessible, modifiable control of the OpenExo system.

Minimum System Requirements¶

Operating System:

Windows 7 or newer
macOS 10.9 or newer

Python Version:

Python 3.2 or newer

Hardware:

Processor: Dual-core (e.g., Intel Core i3 or AMD equivalent)
Memory (RAM): 2 GB minimum (4 GB recommended)
Storage: 100 MB free space (1 GB recommended for large datasets)
Display: 768×1024 resolution minimum

Installation Guide¶

Video Guide: https://youtu.be/VnQeUM2K6Vg

BLE (Bluetooth Low Energy) provides connections between this API and the exoskeletons. We use [Bleak](https://bleak.readthedocs.io/en/latest/) for all BLE operations.

Required Python libraries:

matplotlib
async_tkinter_loop
pygame
numpy
bleak
pandas
scikit-learn
micromlgen
pillow

To install all dependencies, run:

python install_dependencies.py

General Workflow¶

Program control flow diagram:

.. image:: programflow.png
   :alt: Program Control Flow
   :width: 500px

Starting the Application¶

Ensure Bluetooth is enabled on your device.
Open a terminal and navigate to the project directory.
Run:

python3 GUI.py
The GUI will launch with frames for system interaction.

Scanning for Devices¶

Press the Scan button to search for exoskeletons.
If Bluetooth is disabled, an error message will appear.

Navigating the Interface¶

Scan Window: Initiate scanning and connect to devices.
Active Trial: Monitor ongoing trials and sessions.
Biofeedback: Access real-time feedback tools.
Machine Learning: Train and test models for personalized control.
Update Torque: Modify exoskeleton parameters.

Video Example¶

Example Video

Adding a Button¶

To add a button in a frame within the views/ folder:

class YourFrame(tk.Frame):
    def __init__(self, parent):
        super().__init__(parent)
        self.create_widgets()

    def create_widgets(self):
        my_button = tk.Button(
            self,
            text="My New Button",
            command=self.my_button_action
        )
        my_button.pack(pady=10)

    def my_button_action(self):
        print("My New Button was clicked!")

Define button behavior in my_button_action().

Modifying the Graphing¶

Adjust the BasePlot.update_plot method:

def update_plot(self, xValues, yValues, secondY, title):
    max_points = -20  # change to display more/fewer points
    xValues = xValues[max_points:]
    yValues = yValues[max_points:]
    secondY = secondY[max_points:]

Customize axes and titles:

self.ax.set_title(title)
self.ax.set_xlabel("Time")
self.ax.set_ylabel("Value")

Each plot subclass has an animate() method—modify it to change data sourcing or display.

Adding a New Frame¶

1. Create the frame class: Add a new file (e.g., views/newFeature.py)

import tkinter as tk

class NewFeature(tk.Frame):
    def __init__(self, parent, controller):
        super().__init__(parent)
        self.controller = controller

        label = tk.Label(self, text="New Feature Frame", font=("Arial", 24))
        label.pack(pady=20)

        button = tk.Button(
            self,
            text="Go to Scan Window",
            command=lambda: controller.show_frame("ScanWindow")
        )
        button.pack(pady=10)

2. Update the controller: In your main app file, import and register the new frame

from views.newFeature import NewFeature

for F in (ScanWindow, ActiveTrial, UpdateTorque, BioFeedback, MachineLearning, NewFeature):
    # existing frame setup

3. Add navigation: Create a button in any frame to show the new feature

button = tk.Button(
    self,
    text="Go to New Feature",
    command=lambda: controller.show_frame("NewFeature")
)
button.pack(pady=10)

Biofeedback Frame¶

Purpose: Provides visual and auditory feedback during sessions, helping users reach performance goals using real-time sensor data.

Signal Source:

Uses force-sensitive resistors (FSRs) for plantar pressure or estimated ankle power.

Accessed via:

self.master.controller.deviceManager._realTimeProcessor._chart_data

Switch between legs:

_chart_data.leftFsr  or  _chart_data.rightFsr

Modifying the Signal¶

Locate the FSRPlot class in chart.py.
Update the data source variable:

self.master.controller.deviceManager._realTimeProcessor._chart_data

Features and Controls¶

Graphing:

Real-time display of pressure sensor data.
Goal threshold lines when targets are set.

Buttons:

Back: Return to the Active Trial frame.
Left/Right Leg Toggle: Switch graph between legs.
Set Target Value: Input and display training goals.
Reset Target Value: Clear goals.
Mark Trial: Flag events for analysis.
Recalibrate FSRs: Initiate sensor recalibration.