Domenico G. Sorrenti

Rawseeds ground truth collection systems for indoor self-localization and mapping

A trustable and accurate ground truth is a key requirement for benchmarking self-localization and mapping algorithms; on the other hand, collection of ground truth is a complex and daunting task, and its validation is a challenging issue. In this paper we propose two techniques for indoor ground truth collection, developed in the framework of the European project Rawseeds, which are mutually independent and also independent on the sensors onboard the robot. These techniques are based, respectively, on a network of fixed cameras, and on a network of fixed laser scanners. We show how these systems are implemented and deployed, and, most importantly, we evaluate their performance; moreover, we investigate the possible fusion of their outputs.

Omni-directional catadioptric vision for soccer robots

Abstract This paper describes the design of a multi-part mirror catadioptric vision system and its use for self-localization and detection of relevant objects in soccer robots. The mirror and associated algorithms have been used in robots participating in the middle-size league of RoboCup — The World Cup of Soccer Robots.

Omni-Directional Vision with a Multi-part Mirror

This paper presents an omni-directional sensor based on a camera and a mirror generated with a surface of revolution. The requirements the device must fulfill result from its use as the main perception system for the autonomous mobile robots used in F2000 RoboCup competitions. The more relevant requirements which have been pursued are: 1) range sensing in a quite wide region centered around the robot, with good accuracy; 2) sensing around the robot in a given vertical sector, in order to recognize team-mates and adversaries (all robots have a colored marker above a given height); 3) range sensing in a region very close around the robot, with the highest accuracy, to locate and kick the ball. Such requirements have been fulfilled by the design of a mirror built up of three different parts. Each part is devoted to the fulfillment of one requirement. Concerning the first requirement the approach developed is based on the design of a mirrors profile capable to optically compensate the image distortion provided by the mirror profiles commonly used in previous literature. This approach resulted to be similar to a previous work by Hicks and Bajcsy, although independently developed by the authors.

On-Line color calibration in non-stationary environments

In this paper we propose an approach to color classification and image segmentation in non-stationary environments. Our goal is to cope with changing illumination condition by on-line adapting both the parametric color model and its structure/complexity. Other authors used parametric statistics to model color distribution in segmentation and tracking problems, but with a fixed complexity model. Our approach is able to on-line adapt also the complexity of the model, to cope with large variations in the scene illumination and color temperature.

Rawseeds: Building a Benchmarking Toolkit for Autonomous Robotics

Within computer science, autonomous robotics takes the uneasy role of a discipline where the features of both systems (i.e., robots) and their operating environment (i.e., the physical world) conspire to make the application of the experimental scientific method most difficult. This is the reason why much experimental work in robotics is, from the methodological point of view, built on shaky grounds. In particular, scientifically sound benchmarking tools are still largely missing. This chapter starts from Rawseeds, a project focused precisely on benchmarking in robotics, to highlight the reasons for these difficulties and to propose strategies for overcoming some of them. The main result of Rawseeds is a Benchmarking Toolkit: a readily usable instrument to assess and compare algorithms for SLAM, localization, and mapping. Its most innovative aspects include a set of high-quality, validated, multi-sensor datasets, collected both in indoor and in outdoor locations and complemented by ground truth data, and the explicit definition of a set of quantitative performance metrics for the evaluation of algorithms.

Monocular SLAM with Inverse Scaling Parametrization

The recent literature has shown that it is possible to solve t he monocular Simultaneous Localization And Mapping using both undelayed features initialization and an Extedend Kalman Filter. The key concept, to achieve this result, was the introduction of a new parametrization calle d Unified Inverse Depth that produces measurements equations with a high degree of linearity and allows an efficient and accurate modeling of uncertai nties. In this paper we present a monocular EKF SLAM filter based on an altern ative parametrization, i.e., the Inverse Scaling Parametrizati on, characterized by a reduced number of parameters, a more linear measurement model, and a better modeling of features uncertainty for both low and high parallax features. Experiments in simulation demonstrate that the use of the Inverse Scaling solution improves the monocular EKF SLAM filter when compared with the Unified Inverse Depth approach, while experiments on real da ta show the system working as well.

Getting the most from your color camera in a color-coded world

In this paper we present a proposal for setting camera parameters which we claim to give results better matched to applications in color-coded environments then the camera internal algorithms. Moreover it does not require online human intervention, i.e. is automated, and is faster than a human operator. This work applies to situations where the camera is used to extract information from a color-coded world. The experimental activity presented has been performed in the framework of Robocup mid-size rules, with the hypothesis of temporal constancy of light conditions; this work is the necessary first step toward dealing with slow changes, in the time domain, of light conditions.

Background subtraction by combining Temporal and Spatio-Temporal histograms in the presence of camera movement

Background subtraction is the classical approach to differentiate moving objects in a scene from the static background when the camera is fixed. If the fixed camera assumption does not hold, a frame registration step is followed by the background subtraction. However, this registration step cannot perfectly compensate camera motion, thus errors like translations of pixels from their true registered position occur. In this paper, we overcome these errors with a simple, but effective background subtraction algorithm that combines Temporal and Spatio-Temporal approaches. The former models the temporal intensity distribution of each individual pixel. The latter classifies foreground and background pixels, taking into account the intensity distribution of each pixels’ neighborhood. The experimental results show that our algorithm outperforms the state-of-the-art systems in the presence of jitter, in spite of its simplicity.

On the use of inverse scaling in monocular SLAM

Recent works have shown that it is possible to solve the Simultaneous Localization And Mapping problem using an Extended Kalman Filter and a single perspective camera. The principal drawback of these works is an inaccurate modeling of measurement uncertainties, which therefore causes inconsistencies in the filter estimations. A possible solution to proper uncertainty modeling is the Unified Inverse Depth parametrization. In this paper we propose the Inverse Scaling parametrization that still allows an un-delayed initialization of features, while reducing the number of needed parameters and simplifying the measurement model. This novel approach allows a better uncertainty modeling of both low and high parallax features and reduces the likelihood of inconsistencies. Experiments in simulation demonstrate that the use of the Inverse Scaling solution improves the performance of the monocular EKF SLAM filter when compared with the Unified Inverse Depth approach; experiment on real data confirm the applicability of the idea.

MUREA: A MUlti-Resolution Evidence Accumulation Method for Robot Localization in Known Environments

We present MUREA (MUlti-Resolution Evidence Accumulation): a mobile robot localization method for known 2D environments. It is an evidence accumulation method where the complexity is reduced by means of a multi-resolution scheme. The added value of the contribution, in the authors opinion, are 1) the method per se; 2) the capability of the system to accept both raw sensor data as well as independently generated localization estimates; 3) the capability of the system to give out a (less) accurate estimate whenever asked to do so (e.g. before its regular completion), which could be called any-time localization.