A Machine Learning Approach for an Improved Inertial Navigation System Solution

Abstract

The inertial navigation system (INS) is a basic component to obtain a continuous navigation solution in various applications. The INS suffers from a growing error over time. In particular, its navigation solution depends mainly on the quality and grade of the inertial measurement unit (IMU), which provides the INS with both accelerations and angular rates. However, low-cost small micro-electro-mechanical systems (MEMSs) suffer from huge error sources such as bias, the scale factor, scale factor instability, and highly non-linear noise. Therefore, MEMS-IMU measurements lead to drifts in the solutions when used as a control input to the INS. Accordingly, several approaches have been introduced to model and mitigate the errors associated with the IMU. In this paper, a machine-learning-based adaptive neuro-fuzzy inference system (ML-based-ANFIS) is proposed to leverage the performance of low-grade IMUs in two phases. The first phase was training 50% of the low-grade IMU measurements with a high-end IMU to generate a suitable error model. The second phase involved testing the developed model on the remaining low-grade IMU measurements. A real road trajectory was used to evaluate the performance of the proposed algorithm. The results showed the effectiveness of utilizing the proposed ML-ANFIS algorithm to remove the errors and improve the INS solution compared to the traditional one. An improvement of 70% in the 2D positioning and of 92% in the 2D velocity of the INS solution were attained when the proposed algorithm was applied compared to the traditional INS solution.

Keywords: ANFIS; INS; MEMS-IMU; machine learning; navigation; positioning.

Figure 1

Strap down INS block diagram.

Figure 2

The structure of the fuzzy inference system.

Figure 3

The ANFIS’s structure [44].

Figure 4

The block diagram of the training phase of the ML-based-ANFIS showing the model generation process.

Figure 5

The block diagram of the testing phase of the ML-based-ANFIS showing the application of the generated model to the XBOW-IMU and comparing the produced PVA with the reference IMU.

Figure 6

The utilized IMUs mounted on the testbed showing their placement and orientation inside the van.

Figure 7

The 3D gyroscope angular rates with the ML–based–ANFIS (training stage).

Figure 8

The 3D accelerometers with the ML–based–ANFIS (training stage).

Figure 9

The 3D gyroscope angular rates after applying the ML–based–ANFIS (testing stage).

Figure 10

A zoomed–in part of the IMU gyroscope measurements.

Figure 11

The 3D accelerometers with the ML–based–ANFIS (testing stage).

Figure 12

A zoomed–in part of the IMU accelerometers.

Figure 13

Position (Lat, Long, and Alt) components’ comparison.

Figure 14

Velocity ($V_N$, $V_E$, and $V_D$) components’ comparison.

Figure 15

Attitude (roll, pitch, and yaw) angles’ comparison.

Figure 16

Overall trajectory comparison.