Towards persistent indoor appearance-based localization, mapping and navigation using CAT-Graph

The challenge of persistent appearance-based navigation and mapping is to develop an autonomous robotic vision system that can simultaneously localize, map and navigate over the lifetime of the robot. However, the computation time and memory requirements of current appearance-based methods typically scale not only with the size of the environment but also with the operation time of the platform; also, repeated revisits to locations will develop multiple competing representations which reduce recall performance. In this paper we present a solution to the persistent localization, mapping and global path planning problem in the context of a delivery robot in an office environment over a one-week period. Using a graphical appearance-based SLAM algorithm, CAT-Graph, we demonstrate constant time and memory loop closure detection with minimal degradation during repeated revisits to locations, along with topological path planning that improves over time without using a global metric representation. We compare the localization performance of CAT-Graph to openFABMAP, an appearance-only SLAM algorithm, and the path planning performance to occupancy-grid based metric SLAM. We discuss the limitations of the algorithm with regard to environment change over time and illustrate how the topological graph representation can be coupled with local movement behaviors for persistent autonomous robot navigation.

Skype : a communications framework for robotics

This paper describes an architecture for robotic telepresence and teleoperation based on the well known tools ROS and Skype. We discuss how Skype can be used as a framework for robotic communication and can be integrated into a ROS/Linux framework to allow a remote user to not only interact with people near the robot, but to view maps, sensory data, robot pose and to issue commands to the robot’s navigation stack. This allows the remote user to exploit the robot’s autonomy, providing a much more convenient navigation interface than simple remote joysticking.

Reflexive professionalism : reclaiming the voice of authority in shaping the discourses of education policy

The nature and value of ‘professionalism’ has long been contested by both producers and consumers of policy. Most recently, governments have rewritten and redefined professionalism as compliance with externally imposed ‘standards’. This has been achieved by silencing the voices of those who inhabit the professional field of education. This paper uses Foucauldian archaeology to excavate the enunciative field of professionalism by digging through the academic and institutional (political) archive, and in doing so identifies two key policy documents for further analysis. The excavation shows that while the voices of (academic) authority speak of competing discourses emerging, with professional standards promulgated as the mechanism to enhance professionalism, an alternative regime of truth identifies the privileged use of (managerial) voices from outside the field of education to create a discourse of compliance. There has long been a mismatch between the voices of authority on discourses around professionalism from the academic archive and those that count in contemporary and emerging Australian educational policy. In this paper, we counter this mismatch and argue that reflexive educators’ regimes of truth are worthy of attention and should be heard and amplified.

Request for production of documents – UCPR r 222 – manner of compliance

In John Kallinicos Accountants Pty Ltd v Dundrenan Pty Ltd [2009] QDC 141 Irwin DCJ considered the nature of a party’s obligation under r 222 of the Uniform Civil Procedure Rules 1999 (Qld) (UCPR) to produce documents referred to in the parties’ pleadings, particulars or affidavits. The decision examined whether the approach in Belela Pty Ltd v Menzies Excavation Pty Ltd [2005] 2 QdR 230 in relation to disclosure of documents under UCPR r 214 also applied to production of documents under r 222.

Online calibration of stereo rigs for long-term autonomy

Stereo-based visual odometry algorithms are heavily dependent on an accurate calibration of the rigidly fixed stereo pair. Even small shifts in the rigid transform between the cameras can impact on feature matching and 3D scene triangulation, adversely affecting pose estimates and applications dependent on long-term autonomy. In many field-based scenarios where vibration, knocks and pressure change affect a robotic vehicle, maintaining an accurate stereo calibration cannot be guaranteed over long periods. This paper presents a novel method of recalibrating overlapping stereo camera rigs from online visual data while simultaneously providing an up-to-date and up-to-scale pose estimate. The proposed technique implements a novel form of partitioned bundle adjustment that explicitly includes the homogeneous transform between a stereo camera pair to generate an optimal calibration. Pose estimates are computed in parallel to the calibration, providing online recalibration which seamlessly integrates into a stereo visual odometry framework. We present results demonstrating accurate performance of the algorithm on both simulated scenarios and real data gathered from a wide-baseline stereo pair on a ground vehicle traversing urban roads.

Measurement of a PAGAT gel dosimeter by ultrasound computed tomography

In this work we used a 3D quantitative CT ultrasound imaging system to characterise polymer gel dosimeters. The system comprised of two identical 5 MHz 128 element phased-array ultrasound transducers co-axially aligned and submerged in water as a coupling agent. Rotational and translational movement of the gel dosimeter sample between the transducers were performed using a robotic arm. Ultrasound signals were generated and received using an Olympus Omniscan unit. Dose sensitivity of attenuation and time of flight ultrasonic parameters were assessed using this system.

Automatic object segmentation of unstructured scenes using colour and depth maps

This study presents a segmentation pipeline that fuses colour and depth information to automatically separate objects of interest in video sequences captured from a quadcopter. Many approaches assume that cameras are static with known position, a condition which cannot be preserved in most outdoor robotic applications. In this study, the authors compute depth information and camera positions from a monocular video sequence using structure from motion and use this information as an additional cue to colour for accurate segmentation. The authors model the problem similarly to standard segmentation routines as a Markov random field and perform the segmentation using graph cuts optimisation. Manual intervention is minimised and is only required to determine pixel seeds in the first frame which are then automatically reprojected into the remaining frames of the sequence. The authors also describe an automated method to adjust the relative weights for colour and depth according to their discriminative properties in each frame. Experimental results are presented for two video sequences captured using a quadcopter. The quality of the segmentation is compared to a ground truth and other state-of-the-art methods with consistently accurate results.

Nonparametric semantic segmentation for 3D street scenes

In this paper we propose a method to generate a large scale and accurate dense 3D semantic map of street scenes. A dense 3D semantic model of the environment can significantly improve a number of robotic applications such as autonomous driving, navigation or localisation. Instead of using offline trained classifiers for semantic segmentation, our approach employs a data-driven, nonparametric method to parse scenes which easily scale to a large environment and generalise to different scenes. We use stereo image pairs collected from cameras mounted on a moving car to produce dense depth maps which are combined into a global 3D reconstruction using camera poses from stereo visual odometry. Simultaneously, 2D automatic semantic segmentation using a nonparametric scene parsing method is fused into the 3D model. Furthermore, the resultant 3D semantic model is improved with the consideration of moving objects in the scene. We demonstrate our method on the publicly available KITTI dataset and evaluate the performance against manually generated ground truth.

Nonlinear actuator fault detection for small-scale UAS

This paper presents a recursive strategy for online detection of actuator faults on a unmanned aerial system (UAS) subjected to accidental actuator faults. The proposed detection algorithm aims to provide a UAS with the capability of identifying and determining characteristics of actuator faults, offering necessary flight information for the design of fault-tolerant mechanism to compensate for the resultant side-effect when faults occur. The proposed fault detection strategy consists of a bank of unscented Kalman filters (UKFs) with each one detecting a specific type of actuator faults and estimating corresponding velocity and attitude information. Performance of the proposed method is evaluated using a typical nonlinear UAS model and it is demonstrated in simulations that our method is able to detect representative faults with a sufficient accuracy and acceptable time delay, and can be applied to the design of fault-tolerant flight control systems of UASs.

Robots for environmental monitoring : significant advancements and applications

Robotic systems are increasingly being utilised as fundamental data-gathering tools by scientists, allowing new perspectives and a greater understanding of the planet and its environmental processes. Today's robots are already exploring our deep oceans, tracking harmful algal blooms and pollution spread, monitoring climate variables, and even studying remote volcanoes. This article collates and discusses the significant advancements and applications of marine, terrestrial, and airborne robotic systems developed for environmental monitoring during the last two decades. Emerging research trends for achieving large-scale environmental monitoring are also reviewed, including cooperative robotic teams, robot and wireless sensor network (WSN) interaction, adaptive sampling and model-aided path planning. These trends offer efficient and precise measurement of environmental processes at unprecedented scales that will push the frontiers of robotic and natural sciences.

Monotonous trend estimation of deck displacement for automatic landing of rotorcraft UAVs

This paper presents a novel and practical procedure for estimating the mean deck height to assist in automatic landing operations of a Rotorcraft Unmanned Aerial Vehicle (RUAV) in harsh sea environments. A modified Prony Analysis (PA) procedure is outlined to deal with real-time observations of deck displacement, which involves developing an appropriate dynamic model to approach real deck motion with parameters identified through implementing the Forgetting Factor Recursive Least Square (FFRLS) method. The model order is specified using a proper order-selection criterion based on minimizing the summation of accumulated estimation errors. In addition, a feasible threshold criterion is proposed to separate the dominant components of deck displacement, which results in an accurate instantaneous estimation of the mean deck position. Simulation results demonstrate that the proposed recursive procedure exhibits satisfactory estimation performance when applied to real-time deck displacement measurements, making it well suited for integration into ship-RUAV approach and landing guidance systems.

Quantification of ebullitive and diffusive methane release to atmosphere from a water storage

Accurately quantifying total freshwater storage methane release to atmosphere requires the spatial–temporal measurement of both diffusive and ebullitive emissions. Existing floating chamber techniques provide localised assessment of methane flux, however, significant errors can arise when weighting and extrapolation to the entire storage, particularly when ebullition is significant. An improved technique has been developed that compliments traditional chamber based experiments to quantify the storage-scale release of methane gas to atmosphere through ebullition using the measurements from an Optical Methane Detector (OMD) and a robotic boat. This provides a conservative estimate of the methane emission rate from ebullition along with the bubble volume distribution. It also georeferences the area of ebullition activity across entire storages at short temporal scales. An assessment on Little Nerang Dam in Queensland, Australia, demonstrated whole storage methane release significantly differed spatially and throughout the day. Total methane emission estimates showed a potential 32-fold variation in whole-of-dam rates depending on the measurement and extrapolation method and time of day used. The combined chamber and OMD technique showed that 1.8–7.0% of the surface area of Little Nerang Dam is accounting for up to 97% of total methane release to atmosphere throughout the day. Additionally, over 95% of detectable ebullition occurred in depths less than 12 m during the day and 6 m at night. This difference in spatial and temporal methane release rate distribution highlights the need to monitor significant regions of, if not the entire, water storage in order to provide an accurate estimate of ebullition rates and their contribution to annual methane emissions.

The ZomPoc Project. A Large Scale Public ARG Event

An experiment in large scale, live, game design and public performance, bringing together participants from across the creative arts to design, deliver and document a project that was both a cooperative learning experience and an experimental public performance. The four month project, funded by the Edge Digital Centre, culminated into a 24 hour ARG event involving over 100 participants in December 2012. Using the premise of a viral outbreak, young enthusiasts auditioned for the roles of Survivor, Zombie, Medic and Military. The main objective was for the Survivors to complete a series of challenges over 24 hours, while the other characters fulfilled their opposing objectives of interference and sabotage supported by both scripted and free-form scenarios staged in constructed scenes throughout the venues. The event was set in the State Library of Queensland and the Edge Digital Centre who granted the project full access, night and day to all areas including public, office and underground areas. These venues were transformed into cinematic settings full of interactive props and various audio-visual effects. The ZomPoc Project was an innovative experiment in writing and directing a large scale, live, public performance, bringing together participants from across the creative industries. In order to design such an event a number of innovative resources were developed exploiting techniques of game design, theatre, film, television and tangible media production. A series of workshops invited local artists, scientists, technicians and engineers to find new ways of collaborating to create networked artifacts, experimental digital works, robotic props, modular set designs, sound effects and unique costuming guided by an innovative multi-platform script developed by Deb Polson. The result of this collaboration was the creation of innovative game and set props, both atmospheric and interactive. Such works animated the space, presented story clues and facilitated interactions between strangers who found themselves sharing a unique experience in unexpected places.

Dealing with shadows : capturing intrinsic scene appearance for image-based outdoor localisation

In outdoor environments shadows are common. These typically strong visual features cause considerable change in the appearance of a place, and therefore confound vision-based localisation approaches. In this paper we describe how to convert a colour image of the scene to a greyscale invariant image where pixel values are a function of underlying material property not lighting. We summarise the theory of shadow invariant images and discuss the modelling and calibration issues which are important for non-ideal off-the-shelf colour cameras. We evaluate the technique with a commonly used robotic camera and an autonomous car operating in an outdoor environment, and show that it can outperform the use of ordinary greyscale images for the task of visual localisation.

Persistent robot tasking for environmental monitoring through crowd-sourcing

Persistent monitoring of the ocean is not optimally accomplished by repeatedly executing a fixed path in a fixed location. The ocean is dynamic, and so should the executed paths to monitor and observe it. An open question merging autonomy and optimal sampling is how and when to alter a path/decision, yet achieve desired science objectives. Additionally, many marine robotic deployments can last multiple weeks to months; making it very difficult for individuals to continuously monitor and retask them as needed. This problem becomes increasingly more complex when multiple platforms are operating simultaneously. There is a need for monitoring and adaptation of the robotic fleet via teams of scientists working in shifts; crowds are ideal for this task. In this paper, we present a novel application of crowd-sourcing to extend the autonomy of persistent-monitoring vehicles to enable nonrepetitious sampling over long periods of time. We present a framework that enables the control of a marine robot by anybody with an internet-enabled device. Voters are provided current vehicle location, gathered science data and predicted ocean features through the associated decision support system. Results are included from a simulated implementation of our system on a Wave Glider operating in Monterey Bay with the science objective to maximize the sum of observed nitrate values collected.