We present a vision-based navigation system that uses a visual memory to navigate. Such memory corresponds to a topological map of key images created from moving a virtual camera over a model of the real scene. The advantage of our approach is that it provides a useful insight into the navigability of a visual path without relying on a traditional learning stage. During the navigation stage, the robot is controlled by sequentially comparing the images stored in the memory with the images acquired by the onboard camera.The evaluation is conducted on a robotic arm equipped with a camera and the model of the environment corresponds to a top view image of an urban scene.

This paper is about the collaborative use of a far-infrared spectrum human detector and a visible spectrum human detector; the idea is to make collaborate these two detectors of different nature to automatically adapt the human detection whatever the luminosity changes and whatever the infrared emission changes of the scene. Our collaborative approach of detection handles: 1) gradual luminosity changes due, for instance, to the passage from night to day (and vice-versa), 2) sudden luminosity changes due, for instance, to navigation in a forest (when going through a glade in a forest), 3) infrared emission saturation when the global temperature of the scene is very high and does not permit to distinguish human people in infrared. Our approach of detection permits to detect people 24 hours a day and regardless the weather conditions. Furthermore, the proposed approach is relatively fast: it is practically as fast as using one detector alone whereas two are used in the same time.

Nowadays pedestrian detectors are fast, scale-robust and quite efficient. Embedded within a UAV such a detector would open new possibilities. In this paper the very well known HOG detector is adapted for UAV use and a new kind of training dataset is proposed in order to increase the detector's angular robustness. A more appropriate set of detection windows, together with a new detection pipeline, is proposed in order to reduce the search space and consequently reduce the computation time. Tests conducted using the improved detector show significantly better results on aerial images.

Human detection is a very popular field of computer vision. Few works propose a solution for detecting people whatever the camera's viewpoint such as for UAV applications. In this context even state-of-the-art detectors can fail to detect people. We found that the Integral Channel Features detector (ICF) is inoperant in such a context. In this paper, we propose an approach to still benefit from the assets of the ICF while considerably extending the angular robustness during the detection. The main contributions of this work are: a new framework based on the Cluster Boosting Tree and the ICF detector for viewpoint robust human detection; a new training dataset for taking into account the human shape modifications occuring when the pitch angle of the camera changes. We showed that our detector (the PRD) is superior to the ICF for detecting people from complex viewpoints in uncluttered environments and that the computation time of the detector is real-time compatible.

Attitude is one of the most important parameters for a UAV during a flight. Attitude computation methods based vision generally use the horizon line as reference. However, the horizon line becomes an inadequate feature in urban environment. We then propose in this paper an omnidirectional vision system based on straight lines (very frequent in urban environment) that is able to compute the roll and pitch angles. The method consists in finding bundles of horizontal and vertical parallel lines in order to obtain an absolute reference for the attitude computation. We also develop here a new and efficient method for line extraction and bundle of parallel line detection. An original method of horizontal and vertical plane detection is also provided. We show experimental results on different images extracted from video sequences.

Unmanned aerial vehicles (UAVs) are the subject of an increasing interest in many applications. Autonomy is one of the major advantages of these vehicles. It is then necessary to develop particular sensors in order to provide efficient navigation functions. In this paper, we propose a method for attitude computation catadioptric images. We first demonstrate the advantages of the catadioptric vision sensor for this application. In fact, the geometric properties of the sensor permit to compute easily the roll and pitch angles. The method consists in separating the sky from the earth in order to detect the horizon. We propose an adaptation of the Markov random fields for catadioptric images for this segmentation. The second step consists in estimating the parameters of the horizon line thanks to a robust estimation algorithm. We also present the angle estimation algorithm and finally, we show experimental results on synthetic and real images captured from an airplane

Attitude (roll and pitch) is an essential data for the navigation of a UAV. Rather than using inertial sensors, we propose a catadioptric vision system allowing a fast, robust and accurate estimation of these angles. We show that the optimization of a sky/ground partitioning criterion associated with the specific geometric characteristics of the catadioptric sensor provides very interesting results. Experimental results obtained on real sequences are presented and compared with inertial sensor measures

This paper presents a vision system for road traffic surveillance from sequences acquired from an unmanned aerial vehicle (UAV). This UAV is able to follow a path considered as the surveillance area and defined by a set of ordered GPS points. During the navigation of the UAV, the vision system acquires sequences which are treated in real-time in order to detect vehicles. This detection allows to perform a traffic estimation or to track a pointed out vehicle. The detection of vehicles is based on the spatiotemporal grouping of primitives formulated as a normalized cuts problem. A verification step based on the Dempster-Shafer theory is also proposed in order to recognize the vehicles.

This paper deals with an absolute localization paradigm based on the cooperation of an omnidirectional vision system and a low cost panoramic range finder system. These two perception systems used independently by our laboratory provide some complementary data. This association permits us to build a robust sensorial model which integrates an important number of significant primitives. The obtained sensorial model allows one to realize an absolute localization of the mobile robot, in particular, configurations where it is not possible to determine the position with the use of one of the two sensors. First, we present our global perception system. Next, we deal with our sensorial model building approach, and finally we present an absolute-localization method which uses a Dempster-Shafer approach. We analyze the performance of our global absolute localization system on several robot's elementary moves, in an indoor environment.

Presents a system of absolute localization based on the stereoscopic omnidirectional vision. To do it we use an original perception system which allows our omnidirectional vision sensor SYCLOP to move along a rail. The first part of our study deals with the problem of building the sensorial model with the help of the two stereoscopic omnidirectional images. To solve this problem we propose an approach based on the fusion of several criteria which will be made according to Dempster-Shafer rules. The second part is devoted to exploiting this sensorial model to localize the robot thanks to matching the sensorial primitives with the environment map. We analyze the performance of our global absolute localization system on several robot elementary moves, in different environments.

Presents results for state estimation based on noisy observations suffering from two different types of uncertainties. The first uncertainty is a stochastic process with given statistics. The second uncertainty is only known to be bounded, the exact underlying statistics are unknown. State estimation tasks of this kind typically arise in target localization, navigation, and sensor data fusion. A new estimator has been developed, that combines set theoretic and stochastic estimation in a rigorous manner. The estimator is efficient and, hence, well-suited for practical applications. It provides a continuous transition between the two classical estimation, concepts, because it converges to a set theoretic estimator, when the stochastic error goes to zero, and to a Kalman filter, when the bounded error vanishes. In the mixed noise case, the new estimator provides solution sets that are uncertain in a statistical sense.

In this paper we present a method to calibrate our omnidirectional sensor named SYCLOP ("Conic System for Localization and Perception"). Actually this sensor is installing on the mobile robot SARAH, and it is able to rake a panoramic image of the environment. From this image we can localize the mobile robot using natural beacon detection. In order to have a better precision in beacons detection than that obtained in classical applications using this kind of sensors, we bring to the fore the importance of calibration for the whole sensor. We will show to the end of this paper different result obtained in real trajectory.

An omnidirectional range sensor is reported. This active vision sensor combines a CCD camera and a laser diode. We use two methods to obtain the depth of the scene: a calibration method and a least square method. We describe the prototype we made. Experimental results are presented. A comparative test shows that this sensor seems to be as accurate as a laser telemeter but less sensitive to nonalignment. Its results are better than an ultrasonic sensor. Finally, we compare three segmentation algorithms and their results on the set of points given by the sensor.

This paper presents a system of absolute localization based on the cooperation of a stereoscopic omnidirectional vision system and a dead-reckoning system. To do it we use an original perception system which allows our omnidirectional vision sensor SYCLOP to move along a rail. We address the problem of building the sensorial model with the help of the two stereoscopic omnidirectional images. To solve this problem we propose an approach based on the fusion of several criteria which will be made according to Dempster-Shafer rules. We then exploit this sensorial model to localize the robot thanks to matching the sensorial primitives with the environment map. We use the dead-reckoning prediction to decrease the combinatory aspect of the matching algorithm. We analyze the performance of our global absolute localization system on several robots' elementary moves, in an indoor environment.

We present a complete localisation method for mobile robots. We deal with low level image processing, model recognition through an invariant, matching with real noisy goniometrical observations and location estimation through a triangulation technique. We show how useful invariants may be in order to solve the matching problem. Moreover, we point out how invariants are powerful to discriminate features (in images) in order to provide a quick and reliable robot location. Our localisation technique is absolute, i.e. it needs neither a (exterior) position prediction nor a resetting technique. Finally, we present experimental results with a substantial set of real noisy images that demonstrate the reliability of our localisation method.

In this article we present a navigation system allowing a mobile robot to be localized in an indoor environment which is only partially known. This system integrates an environment map updating module allowing the mobile robot to estimate the position of new vertical landmarks along its path. An extended Kalman filter is used on the one hand to estimate the mobile robot position and on the other hand to extract observations which will be used to determine the positions of unlisted landmarks. The integration of new landmarks into the environment global map is managed from the covariance matrix associated with each unlisted landmark. We present the experimental results we have got with SARAH, our mobile robot.

In this paper, we present a new language of communication between cooperative mobile robots. Generally, in order to interact they must exchange various information. In mobile robotics, communication can generally only be performed using HF channel. Originally, our language and protocol are designed to carry out safe communication between several ICAR agents (intelligent cooperative autonomous robots), but they can be used to perform another type of communication. We consider mobile robots which must be able to achieve some tasks given by another one, or a supervisor. We are developing a dynamically extendable (i.e. while robots are performing their missions) language which supports nine conversation types and works with twenty six message types. Because of communications nature (radio link), we use macro messages to reduce network congestion.

We first present a short description of our conic sensor. Furthermore, we pay special attention to the projective invariant, represented by the cross-ratio. We end the paper by presenting a new 1D cross-ratio application that enables us to localise our mobile robot SARAH efficiently. With the method that we propose using, the camera does not need to be calibrated and this vision system, based on an omnidirectional sensor, is a major advantage for the navigation of an autonomous mobile robot.

In this paper we present a dynamic localization system which allows a mobile robot to evolve autonomously in a structured environment. Our system is based on the use of two sensors: an odometer and an omnidirectional vision system which gives a reference in connection with a set of natural beacons. Our navigation algorithm gives a reliable position estimation due to a systematic dynamic resetting. To merge the data obtained we use the extended Kalman filter. Our proposed method allows us to treat efficiently the noise problems linked to the primitive extraction, which contributes to the robustness of our system. Thus, we have developed a reliable and quick navigation system which can deals with the constraints of moving the robots in an industrial environment. We give the experimental results obtained from a mission realized in an a priori known environment.

Mobile robots use actually a combination of internal and external sensors to determine their position and orientation in a path following period. Incremental encoders, gyrometers are generally used to give an approximative estimation of localization. Nevertheless, the cumulative drifts of these internal sensors must be periodically corrected with an exteroceptive sensor. So, the authors present, in this paper, an optical omnidirectional sensor which can give, with an adapted software, an absolute localization. This sensor is made of a CCD video camera associated with a conic shaped reflector; so, a view of a 2 /spl pi/ radian field is available to compute the position of the robot. The authors report the matching algorithm allowing a local observed scene to be replaced in the global navigation area. The authors conduct the accuracy analysis of this global positioning system, and present the experimental results.