State estimation of unmanned vehicles in GPS restricted environment

In this paper, the Extended Kalman Filter(EKF) navigation algorithm method is presented using the data of the inertial measurement unit(IMU) and the encoder aiming at a situation that the external sensors (such as laser, GPS, ultrasonic, etc.) are not available. The IMU mounted on the vehicle is a nine-axis inertial navigation that provides measurements of acceleration and angular velocity. The encoders mounted on the four wheels provides the speed measurement of the each wheel. This experiment is implemented on the (robotic operation system(ROS), and the data of the vehicle is obtained from ROS nodes. Compared with the traditional GPS or laser positioning methods, the EKF navigation algorithm reduces the cost of equipment and can be applied to GPS and lidar positioning failure scenarios. The experiment compares the actual trajectory of the vehicle with the predicted position and orientation of the navigation algorithm, illustrates the effectiveness of the algorithm.