Simultaneous localisation and mapping from natural landmarks using RatSLAM

This paper describes the current state of RatSLAM, a Simultaneous Localisation and Mapping (SLAM) system based on models of the rodent hippocampus. RatSLAM uses a competitive attractor network to fuse visual and odometry information. Energy packets in the network represent pose hypotheses, which are updated by odometry and can be enhanced or inhibited by visual input. This paper shows the effectiveness of the system in real robot tests in unmodified indoor environments using a learning vision system. Results are shown for two test environments; a large corridor loop and the complete floor of an office building.

Probabilistic visual recognition of artificial landmarks for simultaneous localization and mapping

Probabilistic robotics, most often applied to the problem of simultaneous localisation and mapping (SLAM), requires measures of uncertainly to accompany observations of the environment. This paper describes how uncertainly can be characterised for a vision system that locates coloured landmark in a typical laboratory environment. The paper describes a model of the uncertainly in segmentation, the internal camera model and the mounting of the camera on the robot. It =plains the implementation of the system on a laboratory robot, and provides experimental results that show the coherence of the uncertainly model,

Choosing landmarks for risky planning

This work examines the effect of landmark placement on the efficiency and accuracy of risk-bounded searches over probabilistic costmaps for mobile robot path planning. In previous work, risk-bounded searches were shown to offer in excess of 70% efficiency increases over normal heuristic search methods. The technique relies on precomputing distance estimates to landmarks which are then used to produce probability distributions over exact heuristics for use in heuristic searches such as A* and D*. The location and number of these landmarks therefore influence greatly the efficiency of the search and the quality of the risk bounds. Here four new methods of selecting landmarks for risk based search are evaluated. Results are shown which demonstrate that landmark selection needs to take into account the centrality of the landmark, and that diminishing rewards are obtained from using large numbers of landmarks.

A radiological method to determine the accuracy of motion capture marker placement on palpable anatomical landmarks through a shoe

The accuracy of marker placement on palpable surface anatomical landmarks is an important consideration in biomechanics. Although marker placement reliability has been studied in some depth, it remains unclear whether or not the markers are accurately positioned over the intended landmark in order to define the static position and orientation of the segment. A novel method using commonly available X-ray imaging was developed to identify the accuracy of markers placed on the shoe surface by palpating landmarks through the shoe. An anterior–posterior and lateral–medial X-ray was taken on 24 participants with a newly developed marker set applied to both the skin and shoe. The vector magnitude of both skin- and shoe-mounted markers from the anatomical landmark was calculated, as well as the mean marker offset between skin- and shoe-mounted markers. The accuracy of placing markers on the shoe relative to the skin-mounted markers, accounting for shoe thickness, was less than 5mm for all markers studied. Further, when using the developed guidelines provided in this study, the method was deemed reliable (Intra-rater ICCs¼0.50–0.92). In conclusion, the method proposed here can reliably assess marker placement accuracy on the shoe surface relative to chosen anatomical landmarks beneath the skin.

Robust vision-based underwater homing using self-similar landmarks

Next-generation autonomous underwater vehicles (AUVs) will be required to robustly identify underwater targets for tasks such as inspection, localization, and docking. Given their often unstructured operating environments, vision offers enormous potential in underwater navigation over more traditional methods; however, reliable target segmentation often plagues these systems. This paper addresses robust vision-based target recognition by presenting a novel scale and rotationally invariant target design and recognition routine based on self-similar landmarks that enables robust target pose estimation with respect to a single camera. These algorithms are applied to an AUV with controllers developed for vision-based docking with the target. Experimental results show that the system performs exceptionally on limited processing power and demonstrates how the combined vision and controller system enables robust target identification and docking in a variety of operating conditions.

Geometric sensitivity of patient-specific finite element models of the spine to variability in user-selected anatomical landmarks

Software to create individualised finite element (FE) models of the osseoligamentous spine using pre-operative computed tomography (CT) data-sets for spinal surgery patients has recently been developed. This study presents a geometric sensitivity analysis of this software to assess the effect of intra-observer variability in user-selected anatomical landmarks. User-selected landmarks on the osseous anatomy were defined from CT data-sets for three scoliosis patients and these landmarks were used to reconstruct patient-specific anatomy of the spine and ribcage using parametric descriptions. The intra-observer errors in landmark co-ordinates for these anatomical landmarks were calculated. FE models of the spine and ribcage were created using the reconstructed anatomy for each patient and these models were analysed for a loadcase simulating clinical flexibility assessment. The intra-observer error in the anatomical measurements was low in comparison to the initial dimensions, with the exception of the angular measurements for disc wedge and zygapophyseal joint (z-joint) orientation and disc height. This variability suggested that CT resolution may influence such angular measurements, particularly for small anatomical features, such as the z-joints, and may also affect disc height. The results of the FE analysis showed low variation in the model predictions for spinal curvature with the mean intra-observer variability substantially less than the accepted error in clinical measurement. These findings demonstrate that intra-observer variability in landmark point selection has minimal effect on the subsequent FE predictions for a clinical loadcase.

Place recognition with ConvNet landmarks: Viewpoint-robust, condition-robust, training-free

Place recognition has long been an incompletely solved problem in that all approaches involve significant compromises. Current methods address many but never all of the critical challenges of place recognition – viewpoint-invariance, condition-invariance and minimizing training requirements. Here we present an approach that adapts state-of-the-art object proposal techniques to identify potential landmarks within an image for place recognition. We use the astonishing power of convolutional neural network features to identify matching landmark proposals between images to perform place recognition over extreme appearance and viewpoint variations. Our system does not require any form of training, all components are generic enough to be used off-the-shelf. We present a range of challenging experiments in varied viewpoint and environmental conditions. We demonstrate superior performance to current state-of-the- art techniques. Furthermore, by building on existing and widely used recognition frameworks, this approach provides a highly compatible place recognition system with the potential for easy integration of other techniques such as object detection and semantic scene interpretation.

Splintered memories or vivid landmarks? Qualities and organization of traumatic memories with and without PTSD

One hundred and eighty-one students answered a standardized questionnaire on Post-Traumatic Stress Disorder (PTSD): 25 reported trauma(s) and indicated a pattern of after-effects that matched a PTSD symptom profile, whereas 88 indicated trauma(s) but no PTSD symptom profile. Both groups answered a questionnaire addressing the recollective quality, integration and coherence of the traumatic memory that currently affected them most. Participants with a PTSD symptom profile reported more vivid recollection of emotion and sensory impressions. They reported more observer perspective in the memory (seeing themselves 'from the outside'), but no more fragmentation. They also agreed more with the statement that the trauma had become part of their identity, and perceived more thematic connections between the trauma and current events in their lives. The two groups showed different patterns of correlations which indicated different coping styles. Overall, the findings suggest that traumas form dysfunctional reference points for the organization of other personal memories in people with PTSD symptoms, leading to fluctuations between vivid intrusions and avoidance. Copyright © 2003 John Wiley & Sons, Ltd.

The role of visual landmarks in the avian familiar area map - Commentary

The question of whether homing pigeons use visual landmarks for orientation from distant, familiar sites is an unresolved issue in the field of avian navigation. Where evidence has been found, the question still remains as to whether the landmarks are used independent of the map and compass mechanism for orientation that is so important to birds. Recent research has challenged the extent to which experiments that do not directly manipulate the visual sense can be used as evidence for compass-independent orientation. However, it is proposed that extending a new technique for research on vision in homing to include manipulation of the compasses used by birds might be able to resolve this issue. The effect of the structure of the visual sense of the homing pigeon on its use of visual landmarks is also considered.

Anatomical landmarks for transnasal endoscopic skull base surgery.

Resection of midline skull base lesions involve approaches needing extensive neurovascular manipulation. Transnasal endoscopic approach (TEA) is minimally invasive and ideal for certain selected lesions of the anterior skull base. A thorough knowledge of endonasal endoscopic anatomy is essential to be well versed with its surgical applications and this is possible only by dedicated cadaveric dissections. The goal in this study was to understand endoscopic anatomy of the orbital apex, petrous apex and the pterygopalatine fossa. Six cadaveric heads (3 injected and 3 non injected) and 12 sides, were dissected using a TEA outlining systematically, the steps of surgical dissection and the landmarks encountered. Dissection done by the "2 nostril, 4 hands" technique, allows better transnasal instrumentation with two surgeons working in unison with each other. The main surgical landmarks for the orbital apex are the carotid artery protuberance in the lateral sphenoid wall, optic nerve canal, lateral optico-carotid recess, optic strut and the V2 nerve. Orbital apex includes structures passing through the superior and inferior orbital fissure and the optic nerve canal. Vidian nerve canal and the V2 are important landmarks for the petrous apex. Identification of the sphenopalatine artery, V2 and foramen rotundum are important during dissection of the pterygopalatine fossa. In conclusion, the major potential advantage of TEA to the skull base is that it provides a direct anatomical route to the lesion without traversing any major neurovascular structures, as against the open transcranial approaches which involve more neurovascular manipulation and brain retraction. Obviously, these approaches require close cooperation and collaboration between otorhinolaryngologists and neurosurgeons.

Visual SLAM for underwater vehicles using video velocity log and natural landmarks

A visual SLAM system has been implemented and optimised for real-time deployment on an AUV equipped with calibrated stereo cameras. The system incorporates a novel approach to landmark description in which landmarks are local sub maps that consist of a cloud of 3D points and their associated SIFT/SURF descriptors. Landmarks are also sparsely distributed which simplifies and accelerates data association and map updates. In addition to landmark-based localisation the system utilises visual odometry to estimate the pose of the vehicle in 6 degrees of freedom by identifying temporal matches between consecutive local sub maps and computing the motion. Both the extended Kalman filter and unscented Kalman filter have been considered for filtering the observations. The output of the filter is also smoothed using the Rauch-Tung-Striebel (RTS) method to obtain a better alignment of the sequence of local sub maps and to deliver a large-scale 3D acquisition of the surveyed area. Synthetic experiments have been performed using a simulation environment in which ray tracing is used to generate synthetic images for the stereo system