To develop a handheld device which will track the location of an individual relative to a starting point without the use of GPS technologies.
The general method of implementing this objective is outlined in the system block diagram shown below. Accelerometers and gyroscopes will take measurements of the device's acceleration and rotation and output these values to the microprocessor (shown above-right). The processor will then calculate the device's position and compare that position with the position given by the GPS receiver. Depending on certain environment factors, these two position measurements will be weighted against each other and an accurate position produced as a result. The position value is then stored in memory for later use, as well as displayed on the LCD screen.
After having spent two semesters working on this project, we have come to the conclusion that a product like this on the consumer market is still a few years off. Our prototype successfully incorporated all the parts necessary to be a functioning model at a competitive market price; however due to problems with sensor accuracy and a lack of time to further develop our algorithms, our prototype does not successfully use the IMU devices to track position. While tracking is not possible, overall we think this project was a great success. Through it, we learned how to work with modern components (ARM architecture processor) and state-of-the-art components (gyros came to market after we had begun our project!). We also learned about the many complexities involved with inertial navigation that are not obvious immediately, such as drift rates and bias offsets in the sensors. Hopefully in a few years, however, we can begin to see similar products successfully break into the consumer market.