Muhammad AdnanEngineering & designIdeas engineered. Solutions delivered.

Mechanical & CAD

Report-based engineering study

Powered Prosthetic Shoulder: Control & Simulation

An integrated shoulder-drive study linking Simulink motor control, Kalman-filtered IMU data and SolidWorks structural and thermal analysis.

MATLABSimulinkSolidWorksSolidWorks SimulationAMC DriveWare
Simulink block diagram above a plot of input angle, shoulder position and angular velocity. Enlarge image
Original Simulink model and its 90-degree shoulder-position response.
Two original plots show reduced stationary noise and visible lag during rotation. Enlarge image
Stationary and rotating IMU records compare raw angle data with Kalman-filtered output.
CAD views show the articulated arm, shoulder gearbox and placement on a human-form model. Enlarge image
Three original CAD views of the powered arm and shoulder concept.
Von Mises contour includes applied loads, local extrema and a stress legend. Enlarge image
SolidWorks static stress field for the gearbox output shaft.
A SolidWorks thermal contour sits above the original temperature-versus-time graph. Enlarge image
Motor-cage thermal field and temperature history from the 1,200-second study.

Select an image to inspect it. Use the arrow keys to browse the gallery.

Project brief

This project develops a compact powered shoulder concept around a cycloidal transmission and an EC90 brushless motor. Its strongest evidence connects mechanical CAD with sensing and control software. A parameterized MATLAB/Simulink model represents the motor, lower-arm loading and shoulder-position response, allowing motor parameters, payloads and arm dimensions to be varied without rebuilding the complete simulation. The model is used to inspect the voltage, current, angular velocity and torque needed to move the arm toward a commanded position. In parallel, a SparkFun 9DoF IMU is evaluated against an incremental encoder using a pendulum fixture. MATLAB interpolation aligns the two data streams, and a custom Kalman filter combines accelerometer and gyroscope measurements to reduce angle noise and drift. SolidWorks static studies assess the gearbox output shaft and motor cage, while a transient thermal study estimates cage heating under a stated stall-load assumption. The report also documents printed-part tolerancing and an unloaded bidirectional drive demonstration using AMC DriveWare. The simulated arm reaches the reported 90-degree command in approximately two seconds, but loaded precision positioning, complete gearbox dynamics and fatigue life remain unverified. Together, the artifacts show a multidisciplinary prototype workflow with explicit gaps between modeled performance and physical validation.

The engineering challenge

Coordinate a compact shoulder transmission, motor sizing, IMU angle estimation and printed-part strength while distinguishing simulation results from the limited unloaded bench demonstration.

Engineering approach

  1. Size a 39:1 cycloidal transmission around the selected EC90 motor and revise printed-part clearances through fit trials.
  2. Parameterize the motor and lower-arm load in MATLAB and construct a Simulink position-response model.
  3. Compare IMU and encoder pendulum signals, interpolate the records and implement accelerometer–gyroscope Kalman fusion.
  4. Run SolidWorks static shaft and cage studies plus a 1,200-second transient thermal study.
  5. Exercise the shoulder assembly with AMC DriveWare and record the encoder, friction and manufacturing limitations.

Results & observations

39:1Transmission

Selected cycloidal gear reduction with a stated 70% efficiency assumption.

90° in about 2 sModeled position response

Report’s Simulink response to a 90-degree position command with a stated 27 N·m disturbance; not a loaded bench result.

20–40 msFiltered IMU delay

Reported lag in the rotating-IMU example; the source records approximately 260 Hz filtered output.

About 108°C at 20 minThermal simulation

Reported motor-cage maximum under the specified 911 W/m² heat-flux assumption, not a measured operating temperature.

Features & capabilities

  • Parameterized motor model
  • Lower-arm torque subsystem
  • IMU/encoder comparison
  • Kalman sensor fusion
  • Cycloidal gearbox CAD
  • Structural and transient thermal FEA

Software & engineering tools

MATLAB, Simulink, SolidWorks, SolidWorks Simulation, AMC DriveWare, SparkFun 9DoF IMU

More engineering to explore.

Back to all projects