Preliminary Detailed Design
Team Vision for Preliminary Detailed Design PhaseOur goal is to focus on designing a GUI system with a user interface graphical display and to design housings that are able to protect and comfortably hold our sensors. Component analysis will be done, in order to determine which components can form a cohesive device capable of fulfilling all customer needs. While doing so, the risks associated with the components will be analyzed and the components will be priced out in preparation for purchasing.
DeviceTeam P18043 created a functional prototype that we have been using for feasibility testing. By making some small changes to the old hardware/software, we were able to test the EMG sensors, accelerators, gyroscopes, and the serial connection between the device and a computer.
To further refine the device design, less clunky and more straightforward housing designs are being implemented.
IMU Housing: The IMU contains sensitive electronics, and will be in a high use, probable high impact area. Any excess impacts to the IMUs could disrupt the internal geomagnetic sensors, so to prevent this, a more rigid and solid housing is being used.
Arduino Mega Housing: The Arduino Mega contains sensitive electronics, but is in a lower use area and will have a lot more functional components attached to it. As a result,a housing drawing has been made to provide a barrier against impacts, but not hinder any connections contained within the Mega.
DisplayOur current plan for the live display of the data is to use the KST Plot software. KST Plot creates a window that can display multiple plots. The data that the plots show is taken from a data file. Several standard formats are acceptable for this data file including .csv. KST Plot is capable of updating the plots in real time to reflect changes in the data file. We will create several KST Plot sessions, each with a different type of display setup and give the user the option to select their desired display from a separate GUI window. The default option will be to display all data streams.
Feasibility: Analysis, Simulation
Power SubsystemOne component of this design that required feasibility analysis in order to fulfill an engineering requirement was the battery life of the on arm device. This device is meant to run for a full eight hours such that it can run for the entirety of a normal work day. However, due to the strict weight limitation to ensure no tremor mitigation, a light power source was chosen. Based on this, feasibility analysis was done for our arm component. In order to be overly conservative in our estimates, the maximum current of each component will be used as a constant draw to come to conclusions about battery life.
Our battery is a 3000 mAh 3.7 V battery which will be subject to an equivalent max current load of approximately 185.5 mA. This is due to the use of a switching regulator used to step up the supplied voltage to 5 volts. Assuming a very low 50% efficiency, and maximum current draw, the battery capacity should be able to last 8 hours. A breakdown of load values can be seen in the file below titled Arm Power Analysis II.
Our battery system will require monitoring, and after some analysis of parts we determined that the KA2284 Signal Monitor device is the best way to monitor battery life. This component can take in an analog signal and, through the use of low load comparators and voltage division, can display a number of LEDs to reflect the signal level.
Preliminary Electrode PlacementTo fully capture muscle EMG data for the wrist several different motions must be mapped out. The three movements of the wrist the device must be able to capture are Radial/Ulnar Deviation (Left/Right), Flexion/Extension (Up/Down), and Supination/Pronation (Rotating wrist clockwise/counterclockwise).
The first step involved with the electrode placement process was to identify the main muscle groups associated with each of these movements:
|Wrist Motion||Associated Muscle Group|
Flexor Carpi Radialis
Extensor Carpi Radialis Longus
Extensor Carpi Ulnaris
Flexor Carpi Ulnaris
Flexor Carpi Ulnaris
Flexor Carpi Radialis
Extensor Carpi Radialis Longus
Extensor Carpi Radialis Brevis
Extensor Carpi Ulnaris
After mapping out the muscles and cancelling out duplicates, four different electrode sites were determined. These sites are spaced apart from each other on the arm to deter interference and collect EMG signals for all of the motions required of the device. These sites fall along the Extensor Carpi Ulnaris, Flexor Carpi Radialis, Flexor Carpi Ulnaris, and Supinator muscles.
Communication SubsystemIn order to create a wireless communication channel between our arm rig and a computer used for display and analysis, we first needed to ensure that such a feat was possible. The first criterion we determined was what value we would need for a data transmission rate for our device. This was done by first picking a desired sensitivity of 7 characters. This means 7 different ASCII values of 8 bits (ASCII 256) can be sent per each data stream per each cycle. Since we have 16 distinct data-streams (6 for each of 2 IMUs, 1 for each of 4 MyoMuscle Sensors) our data throughput per cycle is 896 bits. We chose to sample each signal at 40 Hz as this value satisfies the Nyquist Criteria for the approximately periodic motion data streams. Due to the nature of MyoMuscle signals being aperiodic envelope functions, we do not need to sample at any specific rate for EMG data. For simplicity's sake, 40 Hz was chosen for all of our signals. This means that our data throughput per second, and therefore our minimum data transmission rate, is 35.84 kb/s. The HC-05 bluetooth device has a maximum data transfer rate of 3 Mb/s. The HC-05 is therefore capable of transmitting data at 83.7 times the minimum required rate. For each set of 16 data points (12 from IMUs, 4 from EMGs), it will take approximately 0.3 ms to transmit the data from the bluetooth device to the computer.
The HC-05 is an I2C Bluetooth Serial Data communicator that was designed to work with Arduino. The HC-05 bluetooth device has a maximum data transfer rate of 3 Mb/s. The HC-05 is therefore capable of transmitting data at 83.7 times the minimum required rate. For each set of 16 data points (12 from IMUs, 4 from EMGs), it will take approximately 0.3 ms to transmit the data from the bluetooth device to the computer. Due to its price, its easy integration, and its ability to satisfy all of our communication needs, we think that this device is a good part to use for wireless data transfer.
Real-Time PlottingUsing the prototype from team P18043, we decided to test out the real-time plotting capabilities of their software. By editing their code, we were able to get multiple signals plotting in real time (EMG & Hand Rotation). This figure only shows the real-time plotting capabilities; the plots are not scaled and aren't even the same units.
The downside of using this real-time plotting technique is that it slows down the sampling frequency of the device. Since the program needs to create/edit plots before the next set of data can be collected, the more plots that are created drop the sampling frequency greatly.
Drawings, Schematics, Flow Charts, SimulationsBlock Diagram
Bill of Material (BOM)BOM Table
Test PlansTest Plans
The previous team (P18043) did not have a very robust graphic user interface to control and manipulate the data acquisition system. For our GUI we will be using a windows executable program. This will allow for a simpler user interface that is easier to use and understand while providing the user with much more control of the device and system.
The main window of the GUI will be small and compact. This keeps the interface simple and provides room to view/manipulate graphs while still seeing the control window.
The Patient Sub-Window is opened from the File tab in the top bar of the main window. From here, you can specify the patient's ID, age, and whether or not they have tremors. These are preliminary information fields that may change later on.
We hope to be able to manipulate the data acquisition system directly from the GUI without editing the actual software. Therefore, from the edit tab, the user will be able to change the Baud Rate and the Com Port. Default values will be set, but if there are changes to the system (IE: the COM port changes), the user will not have to go into the actual software and change values.
Once a device is connected, the baud rate is set to the same value that the device is set to, and the selected com port is correct, then the connection between the device and the computer can be verified (Connection Made). Therefore, it's important to check that the communication between the GUI and device is working properly before we try to collect any data.
Once the device is verified, the user can choose to begin collecting data. While data is being collected, the total recording time will be displayed in the main window.
During recording, the device will send a stream of data to the computer over Bluetooth. This data will be saved into various arrays corresponding to the data (X acceleration, Y rotation, etc). When the user chooses to stop recording, the final time will remain displayed in the main window.
During or after recording, the data can be plotted using the 2 buttons at the bottom of the main window. "Default Plots (All Signals)" will create a new window that displays all the different data plots that have been or are being collected.
The "Add Custom Plot" button will allow the user to create their own plot with signals of their choice. For example, an EMG signal could be placed on the same plot as the hand acceleration in the x-axis.
After the data is collected, the user can save the data into a .csv file or they can just begin recording again.