Robot navigation in sensor space

This thesis investigates the problem of robot navigation using only landmark bearings. The proposed system allows a robot to move to a ground target location specified by the sensor values observed at this ground target posi- tion. The control actions are computed based on the difference between the current landmark bearings and the target landmark bearings. No Cartesian coordinates with respect to the ground are computed by the control system. The robot navigates using solely information from the bearing sensor space. Most existing robot navigation systems require a ground frame (2D Cartesian coordinate system) in order to navigate from a ground point A to a ground point B. The commonly used sensors such as laser range scanner, sonar, infrared, and vision do not directly provide the 2D ground coordi- nates of the robot. The existing systems use the sensor measurements to localise the robot with respect to a map, a set of 2D coordinates of the objects of interest. It is more natural to navigate between the points in the sensor space corresponding to A and B without requiring the Cartesian map and the localisation process. Research on animals has revealed how insects are able to exploit very limited computational and memory resources to successfully navigate to a desired destination without computing Cartesian positions. For example, a honeybee balances the left and right optical flows to navigate in a nar- row corridor. Unlike many other ants, Cataglyphis bicolor does not secrete pheromone trails in order to find its way home but instead uses the sun as a compass to keep track of its home direction vector. The home vector can be inaccurate, so the ant also uses landmark recognition. More precisely, it takes snapshots and compass headings of some landmarks. To return home, the ant tries to line up the landmarks exactly as they were before it started wandering. This thesis introduces a navigation method based on reflex actions in sensor space. The sensor vector is made of the bearings of some landmarks, and the reflex action is a gradient descent with respect to the distance in sensor space between the current sensor vector and the target sensor vec- tor. Our theoretical analysis shows that except for some fully characterized pathological cases, any point is reachable from any other point by reflex action in the bearing sensor space provided the environment contains three landmarks and is free of obstacles. The trajectories of a robot using reflex navigation, like other image- based visual control strategies, do not correspond necessarily to the shortest paths on the ground, because the sensor error is minimized, not the moving distance on the ground. However, we show that the use of a sequence of waypoints in sensor space can address this problem. In order to identify relevant waypoints, we train a Self Organising Map (SOM) from a set of observations uniformly distributed with respect to the ground. This SOM provides a sense of location to the robot, and allows a form of path planning in sensor space. The navigation proposed system is analysed theoretically, and evaluated both in simulation and with experiments on a real robot.

Measuring Visual Consistency in 3D Rendering Systems

One of the major challenges facing a present day game development company is the removal of bugs from such complex virtual environments. This work presents an approach for measuring the correctness of synthetic scenes generated by a rendering system of a 3D application, such as a computer game. Our approach builds a database of labelled point clouds representing the spatiotemporal colour distribution for the objects present in a sequence of bug-free frames. This is done by converting the position that the pixels take over time into the 3D equivalent points with associated colours. Once the space of labelled points is built, each new image produced from the same game by any rendering system can be analysed by measuring its visual inconsistency in terms of distance from the database. Objects within the scene can be relocated (manually or by the application engine); yet the algorithm is able to perform the image analysis in terms of the 3D structure and colour distribution of samples on the surface of the object. We applied our framework to the publicly available game RacingGame developed for Microsoft(R) Xna(R). Preliminary results show how this approach can be used to detect a variety of visual artifacts generated by the rendering system in a professional quality game engine.

The mobile Internet and small business : an exploratory study of needs, uses and adoption with full-adopters of technology

The technological environment in which contemporary small and medium-sized enterprises (SMEs) operate can only be described as dynamic. The exponential rate of technological change, characterised by perceived increases in the benefits associated with various technologies, shortening product life cycles and changing standards, provides for the SME a complex and challenging operational context. The primary aim of this research was to concentrate on those SMEs that had already adopted technology in order to identify their needs for the new mobile data technologies (MDT), the mobile Internet. The research design utilised a mixed approach whereby both qualitative and quantitative data was collected to address the question. Overall, the needs of these SMEs for MDT can be conceptualised into three areas where the technology will assist business practices; communication, eCommerce and security.

3D lung and rib cage reconstruction using low dose CT scans of adolescent idiopathic scoliosis pre and post anterior thoracoscopic correction

Adolescent Idiopathic Scoliosis (AIS) has been associated with reduced pulmonary function believed to be due to a restriction of lung volume by the deformed thoracic cavity. A recent study by our group examined the changes in lung volume pre and post anterior thoracoscopic scoliosis correction using pulmonary function testing (1), however the anatomical changes in ribcage shape and left/right lung volume after thoracoscopic surgery which govern overall respiratory capacity are unknown. The aim of this study was to use 3D rendering from CT scan data to compare lung and ribcage anatomical changes from pre to two years post thoracoscopic anterior scoliosis correction. The study concluded that 3D volumetric reconstruction from CT scans is a powerful means of evaluating changes in pulmonary and thoracic anatomy following surgical AIS correction. Most likely, lung volume changes following thoracoscopic scoliosis correction are multifactorial and affected by changes in height (due to residual growth), ribcage shape, diaphragm positioning, Cobb angle correction in the thoracic spine. Further analysis of the 3D reconstructions will be performed to assess how each of these factors affect lung volume in this patient cohort.

Featureless vehicle-based visual SLAM with a consumer camera

The Simultaneous Localisation And Mapping (SLAM) problem is one of the major challenges in mobile robotics. Probabilistic techniques using high-end range finding devices are well established in the field, but recent work has investigated vision-only approaches. We present an alternative approach to the leading existing techniques, which extracts approximate rotational and translation velocity information from a vehicle-mounted consumer camera, without tracking landmarks. When coupled with an existing SLAM system, the vision module is able to map a 45 metre long indoor loop and a 1.6 km long outdoor road loop, without any parameter or system adjustment between tests. The work serves as a promising pilot study into ground-based vision-only SLAM, with minimal geometric interpretation of the environment.

Single camera vision-only SLAM on a suburban road network

Simultaneous Localization And Mapping (SLAM) is one of the major challenges in mobile robotics. Probabilistic techniques using high-end range finding devices are well established in the field, but recent work has investigated vision only approaches. This paper presents a method for generating approximate rotational and translation velocity information from a single vehicle-mounted consumer camera, without the computationally expensive process of tracking landmarks. The method is tested by employing it to provide the odometric and visual information for the RatSLAM system while mapping a complex suburban road network. RatSLAM generates a coherent map of the environment during an 18 km long trip through suburban traffic at speeds of up to 60 km/hr. This result demonstrates the potential of ground based vision-only SLAM using low cost sensing and computational hardware.

High dynamic range stereo vision for outdoor mobile robotics

We present a technique for high-dynamic range stereo for outdoor mobile robot applications. Stereo pairs are captured at a number of different exposures (exposure bracketing), and combined by projecting the 3D points into a common coordinate frame, and building a 3D occupancy map. We present experimental results for static scenes with constant and dynamic lighting as well as outdoor operation with variable and high contrast lighting conditions.

Hybrid robot control and SLAM for persistent navigation and mapping

For a mobile robot to operate autonomously in real-world environments, it must have an effective control system and a navigation system capable of providing robust localization, path planning and path execution. In this paper we describe the work investigating synergies between mapping and control systems. We have integrated development of a control system for navigating mobile robots and a robot SLAM system. The control system is hybrid in nature and tightly coupled with the SLAM system; it uses a combination of high and low level deliberative and reactive control processes to perform obstacle avoidance, exploration, global navigation and recharging, and draws upon the map learning and localization capabilities of the SLAM system. The effectiveness of this hybrid, multi-level approach was evaluated in the context of a delivery robot scenario. Over a period of two weeks the robot performed 1143 delivery tasks to 11 different locations with only one delivery failure (from which it recovered), travelled a total distance of more than 40km, and recharged autonomously a total of 23 times. In this paper we describe the combined control and SLAM system and discuss insights gained from its successful application in a real-world context.

Properties of photo-cured poly(D,L-lactide) networks

Poly(D,L-lactide) is a degradable polymer with a long history of use in medical applications. It is strong and stiff and degrades over the course of months into lactic acid, a body-own substance. In the field of tissue engineering it is commonly used to fabricate scaffolds. Stereolithography is a high resolution rapid prototyping technique by which designed 3D objects can be built using photo-initiated radical polymerisations. Poly(D,Llactide) (PDLLA) networks can be obtained by photopolymerisation of oligomers functionalised with unsaturated groups. In this work, PDLLA oligomers of varying architectures (arm lengths, numbers of arms) were synthesised and end-functionalised with methacrylate groups. These macromers were photo-crosslinked in solution to yield PDLLA networks of different architectures. The influence of the network architecture on its physical properties was studied.

Influencing the experience of people in urban public places through mobile mediated interactions

This paper introduces our research on influencing the experience of people in urban public places through mobile mediated interactions. Information and communication technology (ICT) devices are sometimes used to create personal space while in public. ICT devices could also be utilised to digitally augment the urban space with non-privacy sensitive data enabling mobile mediated interactions in an anonymous way between collocated strangers. We present what motivates the research on digital augmentations and mobile mediated interactions between unknown urban dwellers, define the research problem that drives this study and why it is significant research in the field of pervasive social networking. The paper illustrates three design interventions enabling social pervasive content sharing and employing pervasive presence, awareness and anonymous social user interaction in urban public places. The paper concludes with an outlook and summarises the research effort.

CAT-SLAM : probabilistic localisation and mapping using a continuous appearance-based trajectory

This paper describes a new system, dubbed Continuous Appearance-based Trajectory Simultaneous Localisation and Mapping (CAT-SLAM), which augments sequential appearance-based place recognition with local metric pose filtering to improve the frequency and reliability of appearance-based loop closure. As in other approaches to appearance-based mapping, loop closure is performed without calculating global feature geometry or performing 3D map construction. Loop-closure filtering uses a probabilistic distribution of possible loop closures along the robot’s previous trajectory, which is represented by a linked list of previously visited locations linked by odometric information. Sequential appearance-based place recognition and local metric pose filtering are evaluated simultaneously using a Rao–Blackwellised particle filter, which weights particles based on appearance matching over sequential frames and the similarity of robot motion along the trajectory. The particle filter explicitly models both the likelihood of revisiting previous locations and exploring new locations. A modified resampling scheme counters particle deprivation and allows loop-closure updates to be performed in constant time for a given environment. We compare the performance of CAT-SLAM with FAB-MAP (a state-of-the-art appearance-only SLAM algorithm) using multiple real-world datasets, demonstrating an increase in the number of correct loop closures detected by CAT-SLAM.

Capping computation time and storage requirements for appearance-based localization with CAT-SLAM

Appearance-based localization is increasingly used for loop closure detection in metric SLAM systems. Since it relies only upon the appearance-based similarity between images from two locations, it can perform loop closure regardless of accumulated metric error. However, the computation time and memory requirements of current appearance-based methods scale linearly not only with the size of the environment but also with the operation time of the platform. These properties impose severe restrictions on longterm autonomy for mobile robots, as loop closure performance will inevitably degrade with increased operation time. We present a set of improvements to the appearance-based SLAM algorithm CAT-SLAM to constrain computation scaling and memory usage with minimal degradation in performance over time. The appearance-based comparison stage is accelerated by exploiting properties of the particle observation update, and nodes in the continuous trajectory map are removed according to minimal information loss criteria. We demonstrate constant time and space loop closure detection in a large urban environment with recall performance exceeding FAB-MAP by a factor of 3 at 100% precision, and investigate the minimum computational and memory requirements for maintaining mapping performance.

The 3D kinematics of the single leg flat and decline squat

Background: Pre-participation screening is commonly used to measure and assess potential intrinsic injury risk. The single leg squat is one such clinical screening measure used to assess lumbopelvic stability and associated intrinsic injury risk. With the addition of a decline board, the single leg decline squat (SLDS) has been shown to reduce ankle dorsiflexion restrictions and allowed greater sagittal plane movement of the hip and knee. On this basis, the SLDS has been employed in the Cricket Australia physiotherapy screening protocols as a measure of lumbopelvic control in the place of the more traditional single leg flat squat (SLFS). Previous research has failed to demonstrate which squatting technique allows for a more comprehensive assessment of lumbopelvic stability. Tenuous links are drawn between kinematics and hip strength measures within the literature for the SLS. Formal evaluation of subjective screening methods has also been suggested within the literature. Purpose: This study had several focal points namely 1) to compare the kinematic differences between the two single leg squatting conditions, primarily the five key kinematic variables fundamental to subjectively assess lumbopelvic stability; 2) determine the effect of ankle dorsiflexion range of motion has on squat kinematics in the two squat techniques; 3) examine the association between key kinematics and subjective physiotherapists’ assessment; and finally 4) explore the association between key kinematics and hip strength. Methods: Nineteen (n=19) subjects performed five SLDS and five SLFS on each leg while being filmed by an 8 camera motion analysis system. Four hip strength measures (internal/external rotation and abd/adduction) and ankle dorsiflexion range of motion were measured using a hand held dynamometer and a goniometer respectively on 16 of these subjects. The same 16 participants were subjectively assessed by an experienced physiotherapist for lumbopelvic stability. Paired samples t-tests were performed on the five predetermined kinematic variables to assess the differences between squat conditions. A Bonferroni correction for multiple comparisons was used which adjusted the significance value to p = 0.005 for the paired t-tests. Linear regressions were used to assess the relationship between kinematics, ankle range of motion and hip strength measures. Bivariate correlations between hip strength measures and kinematics and pelvic obliquity were employed to investigate any possible relationships. Results: 1) Significant kinematic differences between squats were observed in dominant (D) and non-dominant (ND) end of range hip external rotation (ND p = <0.001; D p = 0.004) and hip adduction kinematics (ND p = <0.001; D p = <0.001). With the mean angle, only the non-dominant leg observed significant differences in hip adduction (p = 0.001) and hip external rotation (p = <0.001); 2) Significant linear relationships were observed between clinical measures of ankle dorsiflexion and sagittal plane kinematic namely SLFS dominant ankle (p = 0.006; R2 = .429), SLFS non-dominant knee (p = 0.015; R2 = .352) and SLFS non-dominant ankle (p = 0.027; R2 = .305) kinematics. Only the dominant ankle (p = 0.020; R2 = .331) was found to have a relationship with the decline squat. 3) Strength measures had tenuous associations with the subjective assessments of lumbopelvic stability with no significant relationships being observed. 4) For the non-dominant leg, external rotation strength and abduction strength were found to be significantly correlated with hip rotation kinematics (Newtons r = 0.458 p = 0.049; Normalised for bodyweight: r = 0.469; p = 0.043) and pelvic obliquity (normalised for bodyweight: r = 0.498 p = 0.030) respectively for the SLFS only. No significant relationships were observed in the dominant leg for either squat condition. Some elements of the hip strength screening protocols had linear relationships with kinematics of the lower limb, particularly the sagittal plane movements of the knee and ankle. Strength measures had tenuous associations with the subjective assessments of lumbopelvic stability with no significant relationships being observed; Discussion: The key finding of this study illustrated that kinematic differences can occur at the hip without significant kinematic differences at the knee as a result of the introduction of a decline board. Further observations reinforce the role of limited ankle dorsiflexion range of motion on sagittal plane movement of the hip and knee and in turn multiplanar kinematics of the lower limb. The kinematic differences between conditions have clinical implications for screening protocols that employ frontal plane movement of the knee as a guide for femoral adduction and rotation. Subjects who returned stronger hip strength measurements also appeared to squat deeper as characterised by differences in sagittal plane kinematics of the knee and ankle. Despite the aforementioned findings, the relationship between hip strength and lower limb kinematics remains largely tenuous in the assessment of the lumbopelvic stability using the SLS. The association between kinematics and the subjective measures of lumbopelvic stability also remain tenuous between and within SLS screening protocols. More functional measures of hip strength are needed to further investigate these relationships. Conclusion: The type of SLS (flat or decline) should be taken into account when screening for lumbopelvic stability. Changes to lower limb kinematics, especially around the hip and pelvis, were observed with the introduction of a decline board despite no difference in frontal plane knee movements. Differences in passive ankle dorsiflexion range of motion yielded variations in knee and ankle kinematics during a self-selected single leg squatting task. Clinical implications of removing posterior ankle restraints and using the knee as a guide to illustrate changes at the hip may result in inaccurate screening of lumbopelvic stability. The relationship between sagittal plane lower limb kinematics and hip strength may illustrate that self-selected squat depth may presumably be a useful predictor of the lumbopelvic stability. Further research in this area is required.

Static analysis of executables for collaborative malware detection on Android

Smartphones are getting increasingly popular and several malwares appeared targeting these devices. General countermeasures to smartphone malwares are currently limited to signature-based antivirus scanners which efficiently detect known malwares, but they have serious shortcomings with new and unknown malwares creating a window of opportunity for attackers. As smartphones become host for sensitive data and applications, extended malware detection mechanisms are necessary complying with the corresponding resource constraints. The contribution of this paper is twofold. First, we perform static analysis on the executables to extract their function calls in Android environment using the command readelf. Function call lists are compared with malware executables for classifying them with PART, Prism and Nearest Neighbor Algorithms. Second, we present a collaborative malware detection approach to extend these results. Corresponding simulation results are presented.

Smartphone malware evolution revisited : Android next target?

Smartphones started being targets for malware in June 2004 while malware count increased steadily until the introduction of a mandatory application signing mechanism for Symbian OS in 2006. From this point on, only few news could be read on this topic. Even despite of new emerging smartphone platforms, e.g. android and iPhone, malware writers seemed to lose interest in writing malware for smartphones giving users an unappropriate feeling of safety. In this paper, we revisit smartphone malware evolution for completing the appearance list until end of 2008. For contributing to smartphone malware research, we continue this list by adding descriptions on possible techniques for creating the first malware(s) for Android platform. Our approach involves usage of undocumented Android functions enabling us to execute native Linux application even on retail Android devices. This can be exploited to create malicious Linux applications and daemons using various methods to attack a device. In this manner, we also show that it is possible to bypass the Android permission system by using native Linux applications.