ALBERT

Description: This paper combines the interval analysis tools with the nonlinear model predictive control (NMPC). The NMPC strategy is formulated based on an uncertain dynamic model expressed as nonlinear ordinary differential equations (ODEs). All the dynamic parameters are identified in a guaranteed way considering the various uncertainties on the embedded sensors and the system’s design. The NMPC problem is solved at each time step using validated simulation and interval analysis methods to compute the optimal and safe control inputs over a finite prediction horizon. This approach considers several constraints which are crucial for the system’s safety and stability, namely the state and the control limits. The proposed controller consists of two steps: filtering and branching procedures enabling to find the input intervals that fulfill the state constraints and ensure the convergence to the reference set. Then, the optimization procedure allows for computing the optimal and punctual control input that must be sent to the system’s actuators for the pendulum stabilization. The validated NMPC capabilities are illustrated through several simulations under the DynIbex library and experiments using an inverted pendulum.

Description: The characteristics of bridge pile-group foundation have a significant influence on the dynamic performance of the superstructure. Most of the existing analysis methods for the pile-group foundation impedance take the trait of strong specialty, which cannot be generalized in practical projects. Therefore, a project-oriented numerical solution algorithm is proposed to compute the dynamic impedance of bridge pile-group foundation. Based on the theory of viscous-spring artificial boundary, the derivation and solution of the impedance function are transferred to numerical modeling and harmonic analysis, which can be carried out through the finite element method. By taking a typical pile-group foundation as a case study, the results based on the algorithm are compared with those from existing literature. Moreover, an impact experiment of a real pile-group foundation was implemented, the results of which are also compared with those resulting from the proposed numerical algorithm. Both comparisons show that the proposed numerical algorithm satisfies engineering precision, thus showing good effectiveness in application.

Description: Many smart city and society applications such as smart health (elderly care, medical applications), smart surveillance, sports, and robotics require the recognition of user activities, an important class of problems known as human activity recognition (HAR). Several issues have hindered progress in HAR research, particularly due to the emergence of fog and edge computing, which brings many new opportunities (a low latency, dynamic and real-time decision making, etc.) but comes with its challenges. This paper focuses on addressing two important research gaps in HAR research: (i) improving the HAR prediction accuracy and (ii) managing the frequent changes in the environment and data related to user activities. To address this, we propose an HAR method based on Soft-Voting and Self-Learning (SVSL). SVSL uses two strategies. First, to enhance accuracy, it combines the capabilities of Deep Learning (DL), Generalized Linear Model (GLM), Random Forest (RF), and AdaBoost classifiers using soft-voting. Second, to classify the most challenging data instances, the SVSL method is equipped with a self-training mechanism that generates training data and retrains itself. We investigate the performance of our proposed SVSL method using two publicly available datasets on six human activities related to lying, sitting, and walking positions. The first dataset consists of 562 features and the second dataset consists of five features. The data are collected using the accelerometer and gyroscope smartphone sensors. The results show that the proposed method provides 6.26%, 1.75%, 1.51%, and 4.40% better prediction accuracy (average over the two datasets) compared to GLM, DL, RF, and AdaBoost, respectively. We also analyze and compare the class-wise performance of the SVSL methods with that of DL, GLM, RF, and AdaBoost.

Description: Genetic algorithms (GA’s) are mostly used as an offline optimisation method to discover a suitable solution to a complex problem prior to implementation. In this paper, we present a different application in which a GA is used to progressively adapt the collective performance of an ad hoc collection of devices that are being integrated post-deployment. Adaptive behaviour in the context of this article refers to two dynamic aspects of the problem: (a) the availability of individual devices as well as the objective functions for the performance of the entire population. We illustrate this concept in a video surveillance scenario in which already installed cameras are being retrofitted with networking capabilities to form a coherent closed-circuit television (CCTV) system. We show that this can be conceived as a multi-objective optimisation problem which can be solved at run-time, with the added benefit that solutions can be refined or modified in response to changing priorities or even unpredictable events such as faults. We present results of a detailed simulation study, the implications of which are being discussed from both a theoretical and practical viewpoint (trade-off between saving computational resources and surveillance coverage).

Description: Collision between rigid three-dimensional objects is a very common modelling problem in a wide spectrum of scientific disciplines, including Computer Science and Physics. It spans from realistic animation of polyhedral shapes for computer vision to the description of thermodynamic and dynamic properties in simple and complex fluids. For instance, colloidal particles of especially exotic shapes are commonly modelled as hard-core objects, whose collision test is key to correctly determine their phase and aggregation behaviour. In this work, we propose the Oriented Cuboid Sphere Intersection (OCSI) algorithm to detect collisions between prolate or oblate cuboids and spheres. We investigate OCSI’s performance by bench-marking it against a number of algorithms commonly employed in computer graphics and colloidal science: Quick Rejection First (QRI), Quick Rejection Intertwined (QRF) and a vectorized version of the OBB-sphere collision detection algorithm that explicitly uses SIMD Streaming Extension (SSE) intrinsics, here referred to as SSE-intr. We observed that QRI and QRF significantly depend on the specific cuboid anisotropy and sphere radius, while SSE-intr and OCSI maintain their speed independently of the objects’ geometry. While OCSI and SSE-intr, both based on SIMD parallelization, show excellent and very similar performance, the former provides a more accessible coding and user-friendly implementation as it exploits OpenMP directives for automatic vectorization.

Description: The Dynamic Facility Location problem is a generalization of the classic Facility Location problem, in which the distance metric between clients and facilities changes over time. Such metrics that develop as a function of time are usually called "evolving metrics", thus Dynamic Facility Location can be alternatively interpreted as a Facility Location problem in evolving metrics. The objective in this time-dependent variant is to balance the trade-off between optimizing the classic objective function and the stability of the solution, which is modeled by charging a switching cost when a client’s assignment changes from one facility to another. In this paper, we study the online variant of Dynamic Facility Location. We present a randomized O(logm+logn)-competitive algorithm, where m is the number of facilities and n is the number of clients. In the first step, our algorithm produces a fractional solution, in each timestep, to the objective of Dynamic Facility Location involving a regularization function. This step is an adaptation of the generic algorithm proposed by Buchbinder et al. in their work "Competitive Analysis via Regularization." Then, our algorithm rounds the fractional solution of this timestep to an integral one with the use of exponential clocks. We complement our result by proving a lower bound of Ω(m) for deterministic algorithms and lower bound of Ω(logm) for randomized algorithms. To the best of our knowledge, these are the first results for the online variant of the Dynamic Facility Location problem.

Description: We study the algorithmic complexity of solving subtraction games in a fixed dimension with a finite difference set. We prove that there exists a game in this class such that solving the game is EXP-complete and requires time 2Ω(n), where n is the input size. This bound is optimal up to a polynomial speed-up. The results are based on a construction introduced by Larsson and Wästlund. It relates subtraction games and cellular automata.

Description: This paper proposes a model to evaluate business projects to get into an incubator, allowing to rank them in order of selection priority. The model combines the Momentum method to build prospective scenarios and the AHP-TOPSIS-2N Multiple Criteria Decision Making (MCDM) method to rank the alternatives. Six business projects were evaluated to be incubated. The Momentum method made it possible for us to create an initial core of criteria for the evaluation of incubation projects. The AHP-TOPSIS-2N method supported the decision to choose the company to be incubated by ranking the alternatives in order of relevance. Our evaluation model has improved the existing models used by incubators. This model can be used and/or adapted by any incubator to evaluate the business projects to be incubated. The set of criteria for the evaluation of incubation projects is original and the use of prospective scenarios with an MCDM method to evaluate companies to be incubated does not exist in the literature.

Description: This paper presents an effective stochastic algorithm that embeds a large neighborhood decomposition technique into a variable neighborhood search for solving the permutation flow-shop scheduling problem. The algorithm first constructs a permutation as a seed using a recursive application of the extended two-machine problem. In this method, the jobs are recursively decomposed into two separate groups, and, for each group, an optimal permutation is calculated based on the extended two-machine problem. Then the overall permutation, which is obtained by integrating the sub-solutions, is improved through the application of a variable neighborhood search technique. The same as the first technique, this one is also based on the decomposition paradigm and can find an optimal arrangement for a subset of jobs. In the employed large neighborhood search, the concept of the critical path has been used to help the decomposition process avoid unfruitful computation and arrange only promising contiguous parts of the permutation. In this fashion, the algorithm leaves those parts of the permutation which already have high-quality arrangements and concentrates on modifying other parts. The results of computational experiments on the benchmark instances indicate the procedure works effectively, demonstrating that solutions, in a very short distance of the best-known solutions, are calculated within seconds on a typical personal computer. In terms of the required running time to reach a high-quality solution, the procedure outperforms some well-known metaheuristic algorithms in the literature.

Description: We study cost-sharing games in real-time scheduling systems where the server’s activation cost in every time slot is a function of its load. We focus on monomial cost functions and consider both the case when the degree is less than one (inducing positive congestion effect for the jobs) and when it is greater than one (inducing negative congestion effect for the jobs). For the former case, we provide tight bounds on the price of anarchy, and show that the price of anarchy grows to infinity as a polynomial of the number of jobs in the game. For the latter, we observe that existing results provide constant and tight (asymptotically in the degree of the monomial) bounds on the price of anarchy. We then turn to analyze payment mechanism with arbitrary cost-sharing, that is, when the strategy of a player includes also its payment. We show that our mechanism reduces the price of anarchy of games with n jobs and unit server costs from Θ(n) to 2. We also show that, for a restricted class of instances, a similar improvement is achieved for monomial server costs. This is not the case, however, for unrestricted instances of monomial costs, for which we prove that the price of anarchy remains super-constant for our mechanism. For systems with load-independent activation costs, we show that our mechanism can produce an optimal solution which is stable against coordinated deviations.