scholarly journals Pedestrian Navigation Method Based on Machine Learning and Gait Feature Assistance

Sensors ◽  
2020 ◽  
Vol 20 (5) ◽  
pp. 1530 ◽  
Author(s):  
Zijun Zhou ◽  
Shuqin Yang ◽  
Zhisen Ni ◽  
Weixing Qian ◽  
Cuihong Gu ◽  
...  
Keyword(s):  
Navigation System ◽  
Inertial Navigation System ◽  
Inertial Sensors ◽  
Inertial Navigation ◽  
The Body ◽  
Measurement Unit ◽  
Inertial Measurement ◽  
Pedestrian Navigation ◽  
Gait Feature ◽  
Navigation Method

In recent years, as the mechanical structure of humanoid robots increasingly resembles the human form, research on pedestrian navigation technology has become of great significance for the development of humanoid robot navigation systems. To solve the problem that the wearable inertial navigation system based on micro-inertial measurement units (MIMUs) installed on feet cannot effectively realize its positioning function when the body movement is too drastic to be measured correctly by commercial grade inertial sensors, a pedestrian navigation method based on construction of a virtual inertial measurement unit (VIMU) and gait feature assistance is proposed. The inertial data from different positions of pedestrians’ lower limbs are collected synchronously via actual IMUs as training samples. The nonlinear mapping relationship between inertial information from the human foot and leg is established by a visual geometry group-long short term memory (VGG-LSTM) neural network model, based on which the foot VIMU and virtual inertial navigation system (VINS) are constructed. The VINS experimental results show that, combined with zero-velocity update (ZUPT), the integrated method of error modification proposed in this paper can effectively reduce the accumulation of positioning errors in situations where the gait type exceeds the measurement range of the inertial sensors. The positioning performance of the proposed method is more accurate and stable in complex gait types than that merely using ZUPT.

Aiding Low Cost Inertial Navigation with Building Heading for Pedestrian Navigation

2011 ◽  
Vol 64 (2) ◽  
pp. 219-233 ◽  
Author(s):  
Khairi Abdulrahim ◽  
Chris Hide ◽  
Terry Moore ◽  
Chris Hill
Keyword(s):  
Kalman Filter ◽  
Navigation System ◽  
Inertial Navigation System ◽  
Inertial Sensors ◽  
Low Cost ◽  
Inertial Navigation ◽  
Field Trials ◽  
Basic Knowledge ◽  
Effective Solution ◽  
Pedestrian Navigation

In environments where GNSS is unavailable or not useful for positioning, the use of low cost MEMS-based inertial sensors has paved a way to a more cost effective solution. Of particular interest is a foot mounted pedestrian navigation system, where zero velocity updates (ZUPT) are used with the standard strapdown navigation algorithm in a Kalman filter to restrict the error growth of the low cost inertial sensors. However heading drift still remains despite using ZUPT measurements since the heading error is unobservable. External sensors such as magnetometers are normally used to mitigate this problem, but the reliability of such an approach is questionable because of the existence of magnetic disturbances that are often very difficult to predict. Hence there is a need to eliminate the heading drift problem for such a low cost system without relying on external sensors to give a possible stand-alone low cost inertial navigation system. In this paper, a novel and effective algorithm for generating heading measurements from basic knowledge of the orientation of the building in which the pedestrian is walking is proposed to overcome this problem. The effectiveness of this approach is demonstrated through three field trials using only a forward Kalman filter that can work in real-time without any external sensors. This resulted in position accuracy better than 5 m during a 40 minutes walk, about 0·1% in position error of the total distance. Due to its simplistic algorithm, this simple yet very effective solution is appealing for a promising future autonomous low cost inertial navigation system.

On the Observability of Loosely Coupled Global Positioning System/Inertial Navigation System Integrations With Five Degree of Freedom and Four Degree of Freedom Inertial Measurement Units

2014 ◽  
Vol 136 (2) ◽  
Author(s):  
Jonathan G. Ryan ◽  
David M. Bevly
Keyword(s):  
Global Positioning System ◽  
Navigation System ◽  
Inertial Navigation System ◽  
Inertial Navigation ◽  
Degree Of Freedom ◽  
Measurement Unit ◽  
Positioning System ◽  
Inertial Measurement ◽  
Loosely Coupled ◽  
Global Positioning

This article examines the observability of a modified loosely coupled global positioning system/inertial navigation system (GPS/INS) filter and analyzes the sideslip and attitude estimation capability of the filter. The modified filter is a loosely coupled integration which does not include a pitch rate gyro and which uses GPS course information as a measurement of heading when the vehicle is driving straight. Experimental tests are conducted which show that the modified filter has the same observability characteristics as a standard loosely coupled filter during turning events. The observability of a loosely coupled integration using only a four degree of freedom (DOF) inertial measurement unit (IMU) is also discussed and examined by experiment, as well as the sideslip and roll angle estimation performance. Finally, the error characteristics of the modified loosely coupled integration with the five DOF IMU when the filter is unobservable are studied. Monte Carlo simulations of long periods of straight driving with various sensor qualities are presented to show the worst case attitude errors when the filter is unobservable.

Optimal Sensor Fusion in Redundant Inertial Measurement Unit

2013 ◽  
Vol 433-435 ◽  
pp. 250-253 ◽  
Author(s):  
Xiao Qiang Dai ◽  
Lin Zhao ◽  
Zhen Shi
Keyword(s):  
Sensor Fusion ◽  
Navigation System ◽  
Inertial Measurement Unit ◽  
Inertial Navigation System ◽  
Inertial Navigation ◽  
Measurement Unit ◽  
Inertial Measurement ◽  
Compensation Algorithm ◽  
Drift Compensation

In order to improve the reliability and accuracy of inertial navigation system in some navigation applications, the redundant inertial measurement unit (RIMU) is used. In this study, an optimal sensor fusion is introduced, and drift compensation based on this sensor fusion is presented subsequently. And the sensor fusion and drift compensation are combined into one algorithm. In this algorithm, faulty drift sensors are not isolated, but are compensated. So the redundant inertial navigation system can maintain systems redundancy. An errors model of RIMU was derived firstly, the optimal sensor fusion and the drift compensation algorithm were introduced secondly, and several simulations were carried out and proved effective of the proposed algorithm lastly.

An Improved and Efficient Algorithm for SINS/GPS/Doppler Integrated Navigation Systems

2012 ◽  
Vol 245 ◽  
pp. 323-329 ◽  
Author(s):  
Muhammad Ushaq ◽  
Jian Cheng Fang
Keyword(s):  
Kalman Filter ◽  
Navigation System ◽  
Inertial Navigation System ◽  
Inertial Sensor ◽  
Inertial Navigation ◽  
Measurement Unit ◽  
Navigation Systems ◽  
Integrated Navigation ◽  
Efficient Manner ◽  
Position Errors

Inertial navigation systems exhibit position errors that tend to grow with time in an unbounded mode. This degradation is due, in part, to errors in the initialization of the inertial measurement unit and inertial sensor imperfections such as accelerometer biases and gyroscope drifts. Mitigation to this growth and bounding the errors is to update the inertial navigation system periodically with external position (and/or velocity, attitude) fixes. The synergistic effect is obtained through external measurements updating the inertial navigation system using Kalman filter algorithm. It is a natural requirement that the inertial data and data from the external aids be combined in an optimal and efficient manner. In this paper an efficient method for integration of Strapdown Inertia Navigation System (SINS), Global Positioning System (GPS) and Doppler radar is presented using a centralized linear Kalman filter by treating vector measurements with uncorrelated errors as scalars. Two main advantages have been obtained with this improved scheme. First is the reduced computation time as the number of arithmetic computation required for processing a vector as successive scalar measurements is significantly less than the corresponding number of operations for vector measurement processing. Second advantage is the improved numerical accuracy as avoiding matrix inversion in the implementation of covariance equations improves the robustness of the covariance computations against round off errors.

Sign in / Sign up

Export Citation Format

Share Document