Abstract
The computer-based navigation system computes the object's position, speed, and direction in real-time. In the last decades, many researchers, companies, and industries have been working on improving the existing navigation system due to its vast application in military and civilian activities. Typically, navigation systems are based on integrating inertial navigation systems and global positioning systems using a Bayesian filter, like the Kalman filter. The limitations of the Kalman filter have inspired researchers to consider alternatives based on artificial intelligence. Recently, many types of research have been developed to validate the possibility of using artificial intelligence methods in navigation systems. This chapter aims to review the integration of artificial intelligence techniques in navigation systems.
TopNavigation Systems
The navigation system is a computer-based system embedded in a vehicle or any mobile device that delivers a real-time value of its current location. Recently, many researchers have focused on developing navigation systems due to their importance in civilian and military applications. Modern navigation systems using different electronic sensing devices (sensors) have been developed since the 1960s. These systems combine independent navigation sensors such as inertial measurement units, Doppler radar, and radio position fixed devices to collect the available information to produce a continuous position of the navigated object. Nowadays, microprocessors are integrated with current navigation systems to produce more accurate results (Hasan, Samsudin, Ramli, Azmir, & Ismaeel, 2009). Figure 1 shows the basic structure of navigation systems. The information collected by inertial navigation system (INS) sensors is combined with the position given by the global positioning system (GPS) receiver using a Kalman filter (KF) to give an accurate estimation of the object's position. Usually, a digital map and estimation position are combined to give the object's current position using a map matching technique.
Figure 1. the main structure of a navigation system
Key Terms in this Chapter
Geographic Information System: It produces, manages, and analyzes all kinds of map data.
Extended Kalman Filter: The non-linear version of the Kalman filter linearizes an estimate of the current mean and covariance. In the case of well-defined transition models, the EKF has been used.
Inertial Navigation Unit: It contains an inertial measurement unit (IMU) and computational unit.
Differential Global Positioning System: It utilizes a network of fixed ground-based reference stations to compute and send the difference between the positions given by the GPS satellite system and known fixed positions.
Inertial Measurement Unit: It contains gyroscopes to measure angular rate and accelerometers to measure specific force.
Kalman Filter: It is an algorithm usually used for state estimation problems. It delivers estimates of a system state based on inaccurate and uncertain measurements.
Unscented Kalman Filter: It is a suboptimal non-linear filtration algorithm, however. It uses an unscented transformation (UT) instead of linearization of non-linear equations with the integration of Taylor series expansion.
Map Matching: It matches recorded geographic coordinates to a logical model of real-world objects using a digital map.