Shahram Moafipoor

Pedestrian Tracking and Navigation Using Neural Networks and Fuzzy Logic

The main goal of the research presented here is to develop theoretical foundations and implementation algorithms, which integrate GPS, micro-electro-mechanical inertial measurement unit (MEMS IMU), digital barometer, electronic compass, and human pedometry to provide navigation and tracking of military and rescue ground personnel. This paper discusses the design, implementation and the initial performance analyses of the personal navigator prototype1, with a special emphasis on dead-reckoning (DR) navigation supported by the human locomotion model. To facilitate this functionality, the adaptive knowledge system, based on the Artificial Neural Networks (ANN) and Fuzzy Logic, is trained during the GPS signal reception and used to maintain navigation under GPS-denied conditions. The human locomotion parameters, step frequency (SF) and step length (SL) are estimated during the system calibration period, then the predicted SL, together with the heading information from the compass and gyro, support DR navigation. The current target accuracy of the system is 3-5 m CEP (circular error probable) 50%.

Pedestrian tracking and navigation using an adaptive knowledge system based on neural networks

Abstract The primary objective of the research presented here is to develop theoretical foundations and implementation algorithms, which integrate the Global Positioning System (GPS), micro-electromechanical inertial measurement unit (MEMS IMU), digital barometer, electronic compass, and human pedometry to provide navigation and tracking of military and rescue ground personnel. This paper discusses the design, implementation and the performance analyses of the personal navigator prototype, with a special emphasis on dead-reckoning (DR) navigation supported by the human locomotion model. The adaptive knowledge system, based on the Artificial Neural Networks (ANN), is implemented to support this functionality. The knowledge system is trained during the GPS signal reception and is used to support navigation under GPS-denied conditions. The human locomotion parameters, step frequency (SF) and step length (SL), are extracted from GPS-timed impact switches (step frequency) and GPS/IMU data (step length), respectively, during the system calibration period. SL is correlated with several data types, such as acceleration, acceleration variation, SF, terrain slope, etc. that constitute the input parameters to the ANN-based knowledge system. The ANN-predicted SL, together with the heading information from the compass and gyro, support DR navigation. The current target accuracy of the system is 3–5 m CEP (circular error probable) 50%.

Multi-sensor personal navigator supported by adaptive knowledge based system: Performance assessment

The prototype of a personal navigator, which integrates Global Positioning System (GPS), tactical grade inertial measurement unit (IMU), digital barometer, magnetometer, and human pedometry to support navigation and tracking of military and rescue ground personnel has been developed at The Ohio State University Satellite Positioning and Inertial Navigation (SPIN) Laboratory. This paper discusses the design, implementation and performance assessment of the prototype, with a special emphasis on dead-reckoning (DR) navigation supported by a human locomotion model. The primary components of the human locomotion model are step frequency (SF), extracted from GPS-timed impact micro-switches placed on the shoe soles of the operator, step length (SL), and step direction (SD), both determined by predictive models derived by the adaptive knowledge based system (KBS). SL KBS is based on Artificial Neural Networks (ANN) and Fuzzy Logic (FL), and is trained a priori using sensory data collected by various operators in various environments during GPS signal reception. An additional KBS module, in the form of a Kalman Filter (KF), is used to improve the heading information (SD) available from the magnetometer and gyroscope under GPS-denied conditions, as well as to integrate the DR parameters to reconstruct the trajectory based on SL and SD. The current target accuracy of the system is 3-5 m CEP (circular error probable, 50%). This paper provides a performance analysis in the indoor and outdoor environments for two different operators. The systempsilas navigation limitation in DR mode is tested in terms of time and trajectory length to determine the upper limit of indoor operation before the need for re-calibration.

Real-Time Integrity Monitoring of a Dead Reckoning Personal Navigator Using a Two-Stage Neural Kalman Filter

The basic idea of a dead reckoning personal navigator is to integrate incremental motion information in the forms of step length and step direction over time. Considering that the displacement components are estimated for each step-cycle, it is essential to monitor the integrity of these parameters; otherwise, the error accumulation may render the system unstable. In this paper, a two-stage Kalman Filter (KF) augmented by a neural network is developed to facilitate integrity monitoring. The preliminary results, obtained from several tests performed on simulated and real-word data, indicate an 80% success rate in integrity monitoring.

Multi-Sensor Personal Navigator: System Design and Calibration

This paper presents the current design status and some preliminary calibration/performance analyses of the prototype of a multi-sensor personal navigator currently under development at The Ohio State University. The main purpose of this research project is to develop a theoretical foundation and algorithms which integrate the Global Positioning System (GPS), Micro-electro-mechanical inertial measurement unit (MEMS IMU), barometer and compass to provide seamless position information to support navigation and tracking of ground military and rescue personnel. The system is designed with an open-ended architecture, which would be able to incorporate additional navigation and imaging sensor data, extending the systems operations to the indoor environments. The current target accuracy of the system is at 3–5 m CEP (circular error probable). In the current prototype implementation, the following sensors are integrated in the tightly coupled Extended Kalman Filter (EKF): GPS carrier phase and pseudorange data, Crossbow MEMS IMU400C, PTB220A barometer, and KVH Azimuth 1000 digital compass. This paper focuses on the design architecture of the integrated system and the preliminary performance analysis, with a special emphasis on the navigation during the loss of GPS signal. A brief description of the individual sensors and their calibration is presented, together with the navigation performance of the system of sensors.

Integration of Image-Based and Artificial Intelligence Algorithms: A Novel Approach to Personal Navigation

Navigation systems, such as the Global Positioning System (GPS) and inertial measurement units (IMUs) become miniaturized and cost effective, enabling their fusion in a portable, low-cost navigation device for individual users, supporting predominantly outdoor navigation. This paper presents an unconventional solution designed for indoor–outdoor navigation, based on integration of GPS, IMU, digital barometer, magnetometer compass, and human locomotion model handled by Artificial Intelligence (AI) techniques that form an adaptive knowledge-based system (KBS). KBS is trained during the GPS signal availability, and is used to support navigation under GPS-denied conditions. A complementary technique used in our solution, which supports indoor navigation, is the image-based technique that uses a Flash LADAR sensor. Navigation from 3D Flash LADAR scene reconstruction utilizes the range distance to static features common in images acquired from two separate locations, which allows for triangulating the user’s position. By combining Flash LADAR image with the IMU data, a linear feature-based algorithm that identifies common static features between two images, along with the error estimates, is facilitated. Since the algorithm is based on linear methodologies, it enables rapid processing while generating robust, accurate position and error estimation data. In this paper, system design, as well as a summary of the performance analysis in the mixed indoor–outdoor environments is presented.

Vehicle Classification and Traffic Flow Estimation from Airborne LiDAR/CCD Data

This paper provides a review of a 3-year research program on the feasibility of using airborne LiDAR (Light Detection and Ranging) and imagery collected simultaneously over transportation corridors for estimation of traffic flow parameters such as: (1) vehicle counts, (2) vehicle classification, (3) velocity per vehicle category, and (4) intersection movement patterns. This work is conducted by The National Consortium for Remote Sensing in Transportation-Flows (NCRST-F), led by The Ohio State University, supported by the U.S. Department of Transportation and the National Aeronautics and Space Administration (NASA). The major focus is on improving the efficiency of transportation systems by integration of remotely sensed data with traditional ground data to monitor and manage traffic flows.