Developing methods and algorithms of sensor fusion by IMUs applied to service robotics

The importance of research in Inertial Navigation Systems (INS) has been growing in recent years. Usually the IMU is used in inertial navigation, such as UAV, AGV, AUV, but it is also used in games, human movement reconstruction (the use of sensors in the studies of human movement is now quickly gaining importance as a promising alternative to video capture systems laboratories), entertainment, etc. Often IMU is used in association with GPS or other sensors to estimate trajectory or for navigation as well as localization. In literature, there are many examples using Kalman Filter or EKF for this aim. It can also be considered to use the implementation of an algorithm in the use of the robot. Moreover, when speaking about an inertial platform there is also a Kalman Filter as a good algorithm providing good results. The work described concerned the development of systems and algorithms, or new approaches to existing systems to bring robotics to everyday life and to lower costs of implementation of certain devices in industrial processes, or to review some progresses in the light of improvement of technology. We used the IMUs (Inertial Measurement Unit) and MEMS devices such as accelerometers, gyroscopes, but also temperature and pressure sensors for localization and navigation. Through the use of more accurate sensors and to the growing potential of the new microcontrollers, we have been able to implement algorithms to process and filter data the more quickly and with fewer steps and in some cases to be able to find good solutions at the expense of precision, but in the interest of processing speed. These sensors have been designed as an aid to existing sensors or for new applications such as three-dimensional localization in a building using the pressure or for safety, in industry, eg for the monitoring of movements of a robotic arm. Finally, since usually the inertial navigation uses GPS data to correct inertial data, this excludes the GPS spoofing; in other words that someone deliberately alters the signal in such a way that it provides the same values specifically wrong to hack satellite systems installed in the cars. The IMU used in this work is the iNEMO board, an inertial measurement unit developed by STMicroelectronics. It runs a sophisticated sensor fusion algorithm (attitude heading reference system) to provide static and dynamic orientation and inertial measurements. This 10-DOF inertial system integrates five different sensors and has a size of 4x4 cm.