An intelligent weighted method of calculating angular-velocity in GFSINS

Gyroscope-Free Strap-down Inertial Navigation System (GFSINS) can play the role of inertial measurement and navigation only using accelerometers, which owns wide application foreground. However, the low precision of calculating angular-velocity is always the bottleneck problem of application in reality. To improve it, an intelligent weighted calculating method is designed ably based on a nine accelerometers allocation scheme which can provide two calculated value of angular velocity (integral and extracted values). In every sub-circle of this method, the initial value of integral is updated by use of the calculated weighted value in the previous moment, so the precision of the integral value is improved at the same time. Therefore, a wonderful mutual improving loop of precision is formed. Then the calculating formula of weighted factor is given out by use of genetic algorithm and data fitting method. The simulation result shows that the intelligent weighted method reduces the mean of calculating error 19.7% and the variance two orders of magnitude, compared with traditional calculating methods.