Begoña C. Arrue

An intelligent system for false alarm reduction in infrared forest-fire detection

Forest fires cause many environmental disasters, creating economical and ecological damage as well as endangering peoples lives. Heightened interest in automatic surveillance and early forest-fire detection has taken precedence over traditional human surveillance because the latters subjectivity affects detection reliability, which is the main issue for forest-fire detection systems. In current systems, the process is tedious, and human operators must manually validate many false alarms. Our approach, the False Alarm Reduction system, proposes an alternative real-time infrared-visual system that overcomes this problem. The FAR system consists of applying new infrared-image processing techniques and artificial neural networks (ANNs), using additional information from meteorological sensors and from a geographical information database, taking advantage of the information redundancy from visual and infrared cameras through a matching process, and designing a fuzzy expert rule base to develop a decision function. Furthermore, the system provides the human operator with new software tools to verify alarms.

Computer vision techniques for forest fire perception

This paper presents computer vision techniques for forest fire perception involving measurement of forest fire properties (fire front, flame height, flame inclination angle, fire base width) required for the implementation of advanced forest fire-fighting strategies. The system computes a 3D perception model of the fire and could also be used for visualizing the fire evolution in remote computer systems. The presented system integrates the processing of images from visual and infrared cameras. It applies sensor fusion techniques involving also telemetry sensors, and GPS. The paper also includes some results of forest fire experiments.

Cooperative Large Area Surveillance with a Team of Aerial Mobile Robots for Long Endurance Missions

This paper proposes a distributed approach to solve long duration area surveillance missions with a team of aerial robots, taking into account communication constraints. The system, based on “one-to-one” coordination, minimizes the probability that any event happens in the area without being detected and ensures periodic communication between UAVs. A set of simulations are presented to validate the applicability of the approach and its most relevant features: convergence, robustness with dynamic teams, fault-tolerance and finite information sharing time. It is also shown that “one-to-one” coordination for all the pairs of neighbors allows to obtain an efficient coordination scheme for the whole team to accomplish the area surveillance mission without any central unit.

Control and perception components for autonomous vehicle guidance. Application to the ROMEO vehicles

Abstract This paper presents several control and perception components that implement navigation behaviours in autonomous vehicles. A behaviour-based control architecture integrating these components is also presented. The paper concentrates on autonomous tracking behaviours. In particular, the tracking of explicit paths, moving targets and environmental features are described. These behaviours have been implemented using several different position estimation techniques: dead reckoning; global position estimation using GPS; and position estimation with respect to the environment. The control architecture has been implemented in ROMEO vehicles at the University of Seville. These autonomous vehicles are adaptated from conventional electric vehicles and are currently used for experiments in autonomous navigation and teleoperation in outdoor environments. A short description of the general characteristics of the ROMEO-3R (tricycle) and ROMEO-4R (four wheels) vehicles is presented. The paper includes results from the tracking of environmenal features with proximity sensors, and visual tracking of moving targets.

Fuzzy behaviors combination to control a nonholonomic mobile robot using virtual perception memory

This paper presents the implementation of the combination of fuzzy reactive navigation behaviors using virtual perception memory. Robot control actions are generated by different fuzzy behavior components which cooperate to determine the motion of the vehicle. This approach differs from other methods in the use of a virtual perception memory to make a robot controller more robust with respect to temporary loss of sensorial information. Experimental results of an application to a real robot demonstrate the robustness of the proposed method.

One-to-One Coordination Algorithm for Decentralized Area Partition in Surveillance Missions with a Team of Aerial Robots

This paper presents a decentralized algorithm for area partition in surveillance missions that ensures information propagation among all the robots in the team. The robots have short communication ranges compared to the size of the area to be covered, so a distributed one-to-one coordination schema has been adopted. The goal of the team is to minimize the elapsed time between two consecutive observations of any point in the area. A grid-shape area partition strategy has been designed to guarantee that the information gathered by any robot is shared among all the members of the team. The whole proposed decentralized strategy has been simulated in an urban scenario to confirm that fulfils all the goals and requirements and has been also compared to other strategies.

Smoke monitoring and measurement using image processing: application to forest fires

Smoke detection and monitoring is required for the implementation of advanced forest fire fighting strategies and validation of smoke dispersion models. The latter involve the measurement of smoke column properties. The method proposed in this paper is based on the application of computer-based image processing techniques to visual images taken from fire-spread tests. The method presented involves the application of wavelets and optical flow for fire smoke detection and monitoring. A set of experimental results are reported in the paper, showing the interest of the presented system.

Techniques for reducing false alarms in infrared forest-fire automatic detection systems

This paper presents techniques to reduce false alarms in forest-fire detection systems based on infrared cameras. Several software components to implement these techniques are discussed. The components are based on thresholds, oscillation detection, matching of visual and infrared images, memory, meteorological information, motion, size, shape, solar conditions and location. The paper includes a description of experiments in two forest environments, showing the ability of the proposed techniques to discriminate between forest fires and false alarms. ( 1999 Elsevier Science Ltd. All rights reserved.

Cooperative perimeter surveillance with a team of mobile robots under communication constraints

This paper presents the perimeter surveillance problem using a set of cooperative robots with heterogeneous speed capabilities under communication constraints. The problem is solved using a frequency-based approach. In the proposed path-partition strategy, the robots patrol a segment length related to their own capabilities and interchanges information with its neighbors periodically. An efficient decentralized algorithm is applied to coordinate the robots from local information and decisions. Finally, simulation and experimental results are presented to illustrate the convergence and the robustness of the solution.

Intelligent control of nonholonomic mobile robots with fuzzy perception

This paper is focused on the intelligent control of vehicles with nonholonomic constraints. A new method to compute a fuzzy perception of the environment is presented, dealing with the uncertainties and imprecisions from the sensorial system and taking into account nonholonomic constraints of the vehicle. This fuzzy perception is used, both in the design of each reactive behavior and solving the problem of behavior combination, to implement a fuzzy behavior-based control architecture. Furthermore, teleoperation and planned behaviors, together with their combination with reactive ones, are considered; improving the capabilities of the intelligent control system and its practical application. Experimental results, of an application to control the ROMEO-3R autonomous vehicle, demonstrate the robusmess of the proposed method.