Project Basics:
During my internship at a startup, Haptika, I was assigned a task to develop a hand motion capturing system. This system was to be used in gaming as an alternative for other controllers as well as for other motion related research. Inertial measurement unit (IMU) sensors were to be used in this project. Due to availability of open source libraries and cheap price, I chose to use Invensense's MPU9150 9-axis sensors.
A single MPU9150 sensor has 3-axis accelerometer, gyro-sensor and magnetometer. Magnetometer sensor provides a reference axis to the accelerometer and gyro-sensor readings when they are fused to provide the orientation of MPU9150 sensor. The sensor also has an on-board Digital Motion Processor (DMP) which not only filters the data of three sensor but also fuses it, thus preventing the excessive amount of calculation the controller has to done. Two MPU9150 were used in this system, one for wrist and one for forearm.
The controller used in this project was Arduino Uno. Unity3D was used for the gaming scenes. Initially a character was made in MakeHuman software which was further rigged in Blender3D and its 3D model with its skeletal armature was imported into Unity3D.
Future work involves:
Integrating more sensors for both arms
Implementing wireless communication for better experience
Haptic feedback in the system for real life experience in VR
Changing the controller to Raspberry Pi
Project Supervisor:
Muhammad Nabeel
Members:
Muhammad Shahzaib
Muhammad Ahsan
Huzaifa Sheikh
A single MPU9150 sensor has 3-axis accelerometer, gyro-sensor and magnetometer. Magnetometer sensor provides a reference axis to the accelerometer and gyro-sensor readings when they are fused to provide the orientation of MPU9150 sensor. The sensor also has an on-board Digital Motion Processor (DMP) which not only filters the data of three sensor but also fuses it, thus preventing the excessive amount of calculation the controller has to done. Two MPU9150 were used in this system, one for wrist and one for forearm.
The controller used in this project was Arduino Uno. Unity3D was used for the gaming scenes. Initially a character was made in MakeHuman software which was further rigged in Blender3D and its 3D model with its skeletal armature was imported into Unity3D.
Future work involves:
Integrating more sensors for both arms
Implementing wireless communication for better experience
Haptic feedback in the system for real life experience in VR
Changing the controller to Raspberry Pi
Project Supervisor:
Muhammad Nabeel
Members:
Muhammad Shahzaib
Muhammad Ahsan
Huzaifa Sheikh