Integrated visual perception architecture for robotic clothes perception and manipulation

This thesis proposes a generic visual perception architecture for robotic clothes perception and manipulation. This proposed architecture is fully integrated with a stereo vision system and a dual-arm robot and is able to perform a number of autonomous laundering tasks. Clothes perception and manipulation is a novel research topic in robotics and has experienced rapid development in recent years. Compared to the task of perceiving and manipulating rigid objects, clothes perception and manipulation poses a greater challenge. This can be attributed to two reasons: firstly, deformable clothing requires precise (high-acuity) visual perception and dexterous manipulation; secondly, as clothing approximates a non-rigid 2-manifold in 3-space, that can adopt a quasi-infinite configuration space, the potential variability in the appearance of clothing items makes them difficult to understand, identify uniquely, and interact with by machine. From an applications perspective, and as part of EU CloPeMa project, the integrated visual perception architecture refines a pre-existing clothing manipulation pipeline by completing pre-wash clothes (category) sorting (using single-shot or interactive perception for garment categorisation and manipulation) and post-wash dual-arm flattening. To the best of the author’s knowledge, as investigated in this thesis, the autonomous clothing perception and manipulation solutions presented here were first proposed and reported by the author. All of the reported robot demonstrations in this work follow a perception-manipulation method- ology where visual and tactile feedback (in the form of surface wrinkledness captured by the high accuracy depth sensor i.e. CloPeMa stereo head or the predictive confidence modelled by Gaussian Processing) serve as the halting criteria in the flattening and sorting tasks, respectively. From scientific perspective, the proposed visual perception architecture addresses the above challenges by parsing and grouping 3D clothing configurations hierarchically from low-level curvatures, through mid-level surface shape representations (providing topological descriptions and 3D texture representations), to high-level semantic structures and statistical descriptions. A range of visual features such as Shape Index, Surface Topologies Analysis and Local Binary Patterns have been adapted within this work to parse clothing surfaces and textures and several novel features have been devised, including B-Spline Patches with Locality-Constrained Linear coding, and Topology Spatial Distance to describe and quantify generic landmarks (wrinkles and folds). The essence of this proposed architecture comprises 3D generic surface parsing and interpretation, which is critical to underpinning a number of laundering tasks and has the potential to be extended to other rigid and non-rigid object perception and manipulation tasks. The experimental results presented in this thesis demonstrate that: firstly, the proposed grasp- ing approach achieves on-average 84.7% accuracy; secondly, the proposed flattening approach is able to flatten towels, t-shirts and pants (shorts) within 9 iterations on-average; thirdly, the proposed clothes recognition pipeline can recognise clothes categories from highly wrinkled configurations and advances the state-of-the-art by 36% in terms of classification accuracy, achieving an 83.2% true-positive classification rate when discriminating between five categories of clothes; finally the Gaussian Process based interactive perception approach exhibits a substantial improvement over single-shot perception. Accordingly, this thesis has advanced the state-of-the-art of robot clothes perception and manipulation.

Reactive Attachment Disorder in infants in foster care and associated mental health and cognitive functioning

Background: Reactive attachment disorder (RAD) has been described as one of the least researched and most poorly understood psychiatric disorders (Chaffin et al., 2006). Despite this, given what is known about maltreatment and attachment, it is likely that RAD has profound consequences for child development. Very little is known about the prevalence and stability of RAD symptoms over time. Until recently it has been difficult to investigate the presence of RAD due to limited measures for informing a diagnosis. However this study utilised a new observational tool Method: A cross sectional study design with a one-year follow-up explored RAD symptoms in maltreated infants in Scotland (n=55, age range= 16-62 months) and associated mental health and cognitive functioning. The study utilised the Rating of Inhibited Attachment Behavior Scale (Corval, et al., unpublished 2014) that has recently been developed by experts in the field along side The Disturbances of Attachment Interview (Smyke & Zeanah, 1999). Children were recruited as part of the BeST trial, whereby all infants who came in to the care of the local authority in Glasgow due to child protection concerns were invited to participate. The study sample was representative of the larger pool of data in terms of age, gender, mental health and cognitive functioning. Results: The sample was found to be representative of the population of maltreated children from which it was derived. Prevalence of RAD was found to be 7.3% (n=3, 95% CI [0.43 – 14.17]) at T1, when children are first placed in to foster care. At T2, following one year in improved care conditions, 4.3% (n=2, 95% CI [below 0 – 10.16]) met a borderline RAD diagnosis. Levels of observed RAD symptoms decreased significantly at T2 in comparison to T1 but carer reported symptoms of RAD did not. Children whose RAD symptoms did not improve were found to be significantly older and showed less prosocial behaviour. RAD was associated with some mental health and cognitive difficulties. Lower Verbal IQ and unexpectedly, prosocial behaviour were found to predict RAD symptoms. Conclusions: The preliminary findings have added to the developing understanding of RAD symptoms and associated difficulties however further exploration of RAD in larger samples would be invaluable.