There is an increasing demand for accurate indoor localisation that require minimal infrastructure and wide coverage. All radio technologies commonly used for localisation have trade-offs and there is no clear winner. Accurate systems such as ultra-wideband (UWB) radio, due to their short range of tens of meters, require dense deployment of beacons, incurring huge cost. Such systems out-match the 2.4GHz narrowband technologies with their 10cm accuracy against a few metres with the latter. However, long range technologies such as LoRa, while operating in the 2.4GHz band, can provide a few kilometers of range with fewer beacons in ideal conditions and are therefore cost effective whilst providing greater coverage. We argue in this paper that the desire of having the best of both worlds, i.e., long range, low infrastructure costs, and sub-meter accuracy, can be achieved using embedded machine learning, often referred to as TinyML. To this extent, we run an extensive data collection campaign for two radio technologies i.e., UWB and LoRa, and train machine learning models to improve their localization performance. We then deploy these models on the tiniest of devices powered by ARM Cortex M4 microcontrollers. To the best of our knowledge, we are the first to demonstrate that on-device machine learning can significantly improve localization accuracy. In our case, UWB and LoRa based systems have been shown to improve their localization accuracy by ~20% and ~70% respectively, through learning and essentially calibrating with a more accurate optical camera system without inheriting their weaknesses.

The analysis of athletes’ spatiotemporal data provides actionable insights for strength and conditioning and customized training designs. The identification of unique movements and adjacent match events of team-sport athletes is important. It helps to understand the demands of a match and to advance training programs by improving training specificity. In this paper we present a novel l-length Closed Contiguous sequential pattern mining (LCCspm) algorithm. To validate LCCspm, England Rugby Football League (RFL) Super League players’ movements and Fédération Internationale de Football Association (FIFA) 2018 football world cup events datasets were used. The algorithm was compared with the other existing algorithm (i.e., CCspan). Empirically, the most frequently discovered closed contiguous patterns from RFL were 1-7 length movement patterns while 10-40 length patterns were those discovered in men’s FIFA 2018 world cup. This reflects the duration at which RFL and FIFA football match events usually occur and how data granularity influence results. LCCspm greatly outperforms the CCSpan in terms of scalability, runtime and memory usage. The use of LCCspm instead of CCspan for mining closed contiguous sequences regardless of the length of patterns and size of the database is recommended as it offers timely retrieval of patterns with lesser compute.

In this paper, we revisit the problem online reclustering in clustered shape matching simulations and propose an approach that employs two nonlinear optimizations to create new clusters. The first optimization finds the embedding of particles and clusters into three-dimensional space that minimizes elastic energy. The second finds the optimal location for the new cluster, working in this embedded space. The result is an approach that is more robust in the presence of elastic deformation. We also experimentally verify that our clustered shape matching approach converges as the number of clusters increases, suggesting that our reclustering approach does not change the underlying material properties. Further, we demonstrate that particle resampling is not strictly necessary in our framework allowing us to trivially conserve volume. Finally, we highlight an error in estimating rotations in the original shape-matching work [Müller et al. 2005] that has been repeated in much of the follow up work.

We introduce a new collision proxy for example-based deformable bodies. Specifically, we approximate the deforming geometry as a union of spheres. During pre-computation we perform a sphere packing on the input, undeformed geometry. Then, for each example pose, we move and resize the spheres to approximate the example. During runtime we blend together these positions and radii, using the same skinning weights we use for the geometry. We demonstrate the method on a car crash example, where we achieve an overall speedup of 5--20 times, depending on the resolution of the collision proxy geometry.

In this paper, we incorporate ductile fracture into the clustered shape matching simulation framework for deformable bodies, thus filling a gap in the shape matching literature. Our plasticity and fracture models are inspired by the finite element literature on deformable bodies, but are adapted to the clustered shape matching framework. The resulting approach is fast, versatile, and simple to implement.

Physics-based animation is often used to animate scenes containing destruction of near-rigid, man-made materials. For these applications, the most important visual features are plastic deformation and fracture. Methods based on continuum mechanics model these materials as elastoplastic, and must perform expensive elasticity computations even though elastic deformations are imperceptibly small for rigid materials. We introduce an example-based plasticity model based on linear blend skinning that allows artists to author simulation objects using familiar tools. Dynamics are computed using an unmodified rigid body simulator, making our method computationally efficient and easy to integrate into existing pipelines. We introduce a flexible technique for mapping impulses computed by the rigid body solver to local, example-based deformations. For completeness, our method also supports prescoring based fracture. We demonstrate the practicality of our method by animating a variety of destructive scenes.

Understanding censorship requires performing widespread, continuous measurements "on the ground". Yet, measuring censorship is potentially dangerous, due to the threat of retaliation against citizens who perform measurements. We must balance measurement accuracy, reliability, and scalability with user safety which leads us to the question: Can we design censorship measurements that mitigate risk to the users who consent to perform them? Although it is almost certainly impossible to eliminate risk (or even determine if we have succeeded in doing so), we posit that we may be able to reduce risk with measurement techniques that are difficult to observe or distinguish from innocuous network activity. We observe that surveillance and censorship systems have different goals, and thus certain types of measurement techniques may be able to characterize a censorship system without triggering a surveillance system. We design and implement several techniques for measuring censorship that controlled tests suggest might be less risky than existing methods; we also highlight potential pitfalls, limitations, and avenues for future work.

In this paper, we address clustering and collision detection in the clustered shape matching simulation framework for deformable bodies. Our clustering algorithm is "fuzzy," meaning that it gives particles weighted membership in clusters. These weights are a significant extension to the basic clustered shape matching framework as they are used to divide particle mass among the clusters. We explore several weighting schemes and demonstrate that the choice of weighting scheme gives artists additional control over material behavior. Furthermore, by design our clustering algorithm yields spherical clusters, which not only results in sparse weight vectors, but also exceptionally efficient collision geometry. We further enhance this simple collision proxy by intersecting with half-spaces to allow for even better, yet still simple and computationally efficient, collision proxies. The resulting approach is fast, versatile, and simple to implement.

Abstract Traditional methods for creating dynamic objects and characters from static drawings involve careful tweaking of animation curves and/or simulation parameters. Sprite sheets offer a more drawing‐centric solution, but they do not encode timing information or the logic that determines how objects should transition between poses and cannot generalize outside the given drawings. We present an approach for creating dynamic sprites that leverages sprite sheets while addressing these limitations. In our system, artists create a drawing, deform it to specify a small number of example poses, and indicate which poses can be interpolated. To make the object move, we design a procedural simulation to navigate the pose manifold in response to external or user‐controlled forces. Powerful artistic control is achieved by allowing the artist to specify both the pose manifold and how it is navigated, while physics is leveraged to provide timing and generality. We used our method to create sprites with a range of different dynamic properties. Copyright © 2014 John Wiley & Sons, Ltd.

One means of enforcing Web censorship is to return a block page, which informs the user that an attempt to access a webpage is unsuccessful. Detecting block pages can provide a more complete picture of Web censorship, but automatically identifying block pages is difficult because Web content is dynamic, personalized, and may even be in different languages. Previous work has manually detected and identified block pages, which is difficult to reproduce; it is also time-consuming, which makes it difficult to perform continuous, longitudinal studies of censorship. This paper presents an automated method both to detect block pages and to fingerprint the filtering products that generate them. Our automated method enables continuous measurements of block pages; we found that our methods successfully detect 95% of block pages and identify five filtering tools, including a tool that had not been previously identified "in the wild".

In this paper, we advocate explicit symplectic Euler integration and strain limiting in a shape matching simulation framework. The resulting approach resembles not only previous work on shape matching and strain limiting, but also the recently popular position-based dynamics. However, unlike this previous work, our approach reduces to explicit integration under small strains, but remains stable in the presence of non-linearities.

We present a straightforward, easy-to-implement, point-based approach for animating elastoplastic materials. The core idea of our approach is the introduction of embedded space —the least-squares best fit of the material's rest state into three dimensions. Nearest-neighbor queries in the embedded space efficiently update particle neighborhoods to account for plastic flow. These queries are simpler and more efficient than remeshing strategies employed in mesh-based finite element methods. We also introduce a new estimate for the volume of a particle, allowing particle masses to vary spatially and temporally with fixed density. Our approach can handle simultaneous extreme elastic and plastic deformations. We demonstrate our approach on a variety of examples that exhibit a wide range of material behaviors.

Traditional methods for creating dynamic objects and characters from static drawings involve careful tweaking of animation curves and/or simulation parameters. Sprite sheets offer a more drawing-centric solution, but they do not encode timing information or the logic that determines how objects should transition between poses and cannot generalize outside the given drawings. We present an approach for creating dynamic sprites that leverages sprite sheets while addressing these limitations. In our system, artists create a drawing, deform it to specify a small number of example poses, and indicate which poses can be interpolated. To make the object move, we design a procedural simulation to navigate the pose manifold in response to external or user-controlled forces. Powerful artistic control is achieved by allowing the artist to specify both the pose manifold and how it is navigated, while physics is leveraged to provide timing and generality. We used our method to create sprites with a range of different dynamic properties.

Embedding high-resolution surface geometry in coarse control meshes is a standard approach to achieving high-quality computer animation at low computational expense. In this paper we present an effective, automatic method for generating such control meshes. The resulting high-quality, tetrahedral meshes enclose and approximate an input surface mesh, avoiding extrapolation artifacts and ensuring that the resulting coarse volumetric meshes are adequate collision proxies. Our approach comprises three steps: we begin with a tetrahedral mesh built from the body-centered cubic lattice that tessellates the bounding box of the input surface; we then perform a sculpting phase that carefully removes elements from the lattice; and finally a variational vertex adjustment phase iteratively adjusts vertex positions to more closely approximate the surface geometry. Our approach provides explicit trade-offs between mesh quality, resolution, and surface approximation. Our experiments demonstrate the technique can be used to build high-quality meshes appropriate for simulations within games.

Body morphology is thought to have heavily influenced the evolution of neural architecture. However, the extent of this interaction and its underlying principles are largely unclear. To help us elucidate these principles, we examine the artificial evolution of a hypothetical nervous system embedded in a fish-inspired animat. The aim is to observe the evolution of neural structures in relation to both body morphology and required motor primitives. Our investigations reveal that increasing the pressure to evolve a wider range of movements also results in higher levels of neural symmetry. We further examine how different body shapes affect the evolution of neural structure; we find that, in order to achieve optimal movements, the neural structure integrates and compensates for asymmetrical body morphology. Our study clearly indicates that different parts of the animat - specifically, nervous system and body plan - evolve in concert with and become highly functional with respect to the other parts. The autonomous emergence of morphological and neural computation in this model contributes to unveiling the surprisingly strong coupling of such systems in nature.

We develop an integrated set of gaits and skills for a physics-based simulation of a quadruped. The motion repertoire for our simulated dog includes walk, trot, pace, canter, transverse gallop, rotary gallop, leaps capable of jumping on-and-off platforms and over obstacles, sitting, lying down, standing up, and getting up from a fall. The controllers use a representation based on gait graphs, a dual leg frame model, a flexible spine model, and the extensive use of internal virtual forces applied via the Jacobian transpose. Optimizations are applied to these control abstractions in order to achieve robust gaits and leaps with desired motion styles. The resulting gaits are evaluated for robustness with respect to push disturbances and the traversal of variable terrain. The simulated motions are also compared to motion data captured from a filmed dog.

An agent that deviates from a usual or previous course of action can be said to display novel or varying behaviour. Novelty of behaviour can be seen as the result of real or apparent randomness in decision making, which prevents an agent from repeating exactly past choices. In this paper, novelty of behaviour is considered as an evolutionary precursor of the exploring skill in reward learning, and conservative behaviour as the precursor of exploitation. Novelty of behaviour in neural control is hypothesised to be an important factor in the neuro-evolution of operant reward learning. Agents capable of varying behaviour, as opposed to conservative, when exposed to reward stimuli appear to acquire on a faster evolutionary scale the meaning and use of such reward information. The hypothesis is validated by comparing the performance during evolution in two environments that either favour or are neutral to novelty. Following these findings, we suggest that neuro-evolution of operant reward learning is fostered by environments where behavioural novelty is intrinsically beneficial, i.e. where varying or exploring behaviour is associated with low risk.

The bacterium Escherichia coli has the capacity to respond to a wide range of environmental inputs, which have the potential to change suddenly and rapidly. Although the functions of many of its signal transduction and gene regulation networks have been identified, E.Coli's capacity for perceptual categorization, especially for discrimination between complex temporal patterns of chemical inputs, has been experimentally neglected. Real-time computations on time-varying inputs can be undertaken by a system possessing a high dimensional analog fading memory, i.e. a liquid-state machine (LSM). For example, the cortical microcolumn is hypothesized to be a LSM. A model of the gene regulation network (GRN) of E.Coli was assessed for its LSM properties for a range of increasingly complex stimuli. Cooperativity between transcription factors (TFs) is necessary for complex temporal discriminations. However, the low recurrence within the GRNs autonomous dynamics decreases its capacity for a rich fading memory, and hence for integrating temporal sequence information. We conclude that coupling of the GRN with signal transduction networks possessing cross-talk, and with metabolic networks is expected to increase the extent of non-autonomous recurrence and hence to facilitate enhanced LSM properties.