Hide menu

Research areas

Currently, most of my research is related to the following topics.

Underwater monitoring and surveillance

During the last decades a range of new signal processing theories and methods for target detection and tracking, such as random finite set filters, adaptive waveform design, and track-before-detect filters, have been developed. Along the same line of research, methods for sensor management and distributed sensor fusion in sensor networks subjected to energy, computational, and communication constraints, have been developed. These methods have successfully been applied in various terrestrial radar systems and sensor networks. However, due to the complex acoustic environment the application of these theories and methods in sub-surface monitoring and surveillance systems remains a challenge. This is especially the case for systems that are to operate in littoral waters, such as the Swedish coast. The shallow depths, surface heating and cooling, salinity changes, dense civil traffic, etc., result in an ever-changing acoustic environment; an environment to which the theories and methods must be adapted and tuned. Generally, the complex acoustic environment also limits the communication capabilities within underwater sensor networks to a point beyond that imposed on terrestrial sensor networks. This implies that the employed distributed target detection and tracking methods have to function with a minimum of information exchange, and new communication strategies that make dual use of the communication signals both for sensing and information exchange, are necessary.

Within the Complex Acoustic Surveillance and Tracking (COAST) project , I in cooperation with the Dept. of Naval Systems & Underwater Technology, Swedish Defence Research Agency (FOI), and Saab Dynamics, research and develop methods to adapt current state-of-the-art target detection and tracking methods to the complex acoustic environment encountered by sub-surface systems.

Illustration of a signals of opportunity based array shapes estimation setup. From the sound emitted by a ship passing by the arrays time difference of arrival estimates are calculated and used to estimate the location and shape of the arrays.
Initial project results about signals of opportunity based geometry calibration of hydrophone arrays are presented in the paper:

Inertial and magnetic-field sensor arrays

Today, thanks to development of the micro-electromechanical-systems (MEMS) technology, miniaturized accelerometers, gyroscopes, and magnetometers can be produced at unprecedented volumes and at low price. Unfortunately, the performance of these ultra-low-cost sensors is too poor to enable accuracte and reliable localization in many applications. However, by capitalizing on the small size, low price, and low power consumption of the sensors it is now feasible to construct arrays with hundreds of sensing elements and fuse their measurement to create "super sensors" with unprecedented performance-to-price ratios;
In-house developed smart inertial sensor array with 288 sensing elements. The range, sampling time, and bandwidth of the individual sensors can be controlled via the software of the embedded processor. This enables the array to adapt to the usage conditions.
However, before these "super sensors" can become real, several challanges related to: how to fuse the information from the sensors; how to calibrate the sensor arrays; and how to automatically adapt the sensors settings to different usage conditions; needs to be solved. I together my colleagues have looked at some of these challanges in the following papers: With an array of spatially distributed magnetometers it is possible to instantaneously estimate the Jacobian of the local magnetic field. This enables more accurate and robust magnetic-field based simultaneous localization and mapping (SLAM); a technique frequently used for indoor localization of robots and pedestrians. More importantly, it enables magnetic-field odometry. That is, the speed of the array can be estimated without prior mapping of the magnetic field anomalies. This is very interesting as a classical inertial navigation system have a cubic drift over time due to the combined effects of double integrating acceleration and gravity leakage, whereas an inertial navigation system combined with an array of magnetometers potentially only has a linear drift over time; a prospective game changer for certain indoor navigation systems. An initial experimental evaluation of the concept of magnetic-field odometry is presented in the paper: As the initial results were promising, the research is continued within the Security-Link project: Magnetic-Field based Speed Aided Inertial Navigation

Insurance telematics

Insurance telematics is a disruptive technology that is expected to reform the vehicle insurance industry. Based on sensor data, the traditional measures for calculating the insurance premium are complemented to determine a fee that more accurately predicts the risk profile of the policyholder. From an signal processing point of view, there are several insurance telematics challenges that have to be tackled. It is about consistently, from low quality data, extracting relevant figures of merit (FoMs) like number of harsh braking, speeding, heavy cornering, trip smoothness, etc, and then to transform these FoMs into a valid measure, or score, that determines the risk profile of the insurance customer.

During the last 4 years I, together with Prof. P. Handel (KTH) and J. Wahlstrom (KTH), have conducted research on how the smartphone can be used as a measurement probe in insurance telematics applications. The main focus has been on developing signal processing methods that enables (reliable) calculation of driver behavior FoM from the often poor data provided by the smartphone sensors. Some of our most recent publications on the topic are:
Isaac Skog

Associate Professor in Automatic Control,
Docent in Signal Processing

(Swedish: Universitetslektor i reglerteknik)

Phone:
+46 708186805
E-mail:
isaac.skog_at_liu.se
Address:
Dept. of Electrical Engineering
Linköping University
SE-581 83 Linköping
Sweden
Visiting Address:
Campus Valla
Building B
Room 2A:526


Page responsible: Isaac Skog
Last updated: 2019-10-24