Search

Predicting smartphone users location with WiFi fingerprints has been a popular research topic recently. In this work, we propose two novel deep learning-based models, the convolutional mixture density recurrent neural network and the VAE-based semi-supervised learning model. The convolutional mixture density recurrent neural network is designed for path prediction, in which the advantages of convolutional neural networks, recurrent neural networks and mixture density networks are combined. Further, since most of real-world datasets are not labeled, we devise the VAE-based model for the semi-supervised learning tasks. In order to test the proposed models, we conduct the validation experiments on the real-world datasets. The final results verify the effectiveness of our approaches and show the superiority over other existing methods., Comment: 11 pages, 10 figures

Vital records are rich of meaningful historical data concerning city as well as countryside inhabitants that can be used, among others, to study former populations and then reveal the social, economic and demographic characteristics of those populations. However, these studies encounter a main difficulty for collecting the data needed since most of these records are scanned documents that need a manual transcription step in order to gather all the data and start exploiting it from a historical point of view. This step consequently slows down the historical research and is an obstacle to a better knowledge of the population habits depending on their social conditions. Therefore in this paper, we present a modular and self-sufficient analysis pipeline using state-of-the-art algorithms mostly regardless of the document layout that aims to automate this data extraction process.

Recognizing user location with WiFi fingerprints is a popular approach for accurate indoor positioning problems. In this work, our goal is to interpret WiFi fingerprints into actual user locations. However, WiFi fingerprint data can be very high dimensional in some cases, we need to find a good representation of the input data for the learning task first. Otherwise, using neural networks will suffer from severe overfitting. In this work, we solve this issue by combining the Information Bottleneck method and Variational Inference. Based on these two approaches, we propose a Variational Information Bottleneck model for accurate indoor positioning. The proposed model consists of an encoder structure and a predictor structure. The encoder is to find a good representation in the input data for the learning task. The predictor is to use the latent representation to predict the final output. To enhance the generalization of our model, we also adopt the Dropout technique for each hidden layer of the decoder. We conduct the validation experiments on a real-world dataset. We also compare the proposed model to other existing methods so as to quantify the performances of our method.

Analyzing human mobility with geo-location data collected from smartphones has been a hot research topic in recent years. In this paper, we attempt to discover daily mobile patterns using the GPS data. In particular, we view this problem from a probabilistic perspective. A non-parametric Bayesian modeling method, the Infinite Gaussian Mixture Model (IGMM) is used to estimate the probability density of the daily mobility. We also utilize the Kullback-Leibler (KL) divergence as the metrics to measure the similarity of different probability distributions. Combining the IGMM and the KL divergence, we propose an automatic clustering algorithm to discover mobility patterns for each individual user. Finally, the effectiveness of our method is validated on the real user data collected from different real users.

Augmented Reality (AR) is now more and more widespread in many application fields. Thanks to the new progresses in computer power, AR can now be used widely in education without any expensive additional devices. In this paper is presented the feedback of an experimental protocol using an AR application as an additional support for a History lesson in secondary schools. Even if the technical part has a lead role in the student experience, the most challenging issue is related to the choice of the teaching lesson as itself which must fit several, sometimes contradictory, requirements.

The use of technologies such as Virtual Reality (VR) and Augmented Reality (AR) is more and more widespread in our everyday life and is now mature enough to be used in critical medical or aerospace applications. The main issue is then to provide more adaptive abilities to VR/AR applications in developing them using human concerns as primary element. To that way, beyond the classical HITL concept this paper proposes an integrative point of view for virtual environment design based on the concept of Bio-CPS. The goal of this paper is to present Bio-CPS related concepts and to sketch some guidelines for future virtual environments design focusing on specific application fields.

Reactive multi-agent systems are increasingly applied to problem solving, modelling and simulation. Among the benefits of this approach, there is the capability to solve complex problems, whereas maintaining functional and conceptual simplicity of involved entities. Many examples have shown that approaches based on multi-agent systems are effective in solving complex problems such as life simulation and multi-robot cooperation for instance. Side-by-side vehicle platoon systems, which are studied in this article, can be adopted as a representative example of this class of applications. Based on preceding works, this study presents a local platoon control approach with obstacle avoidance. The proposed vehicle decision-making process is considered as a multi-agent system, the agents of which make collectively the best decision considering the perceived constraints and the preceding vehicle position and speed.

International audience; The monitoring and the surveillance of industrial and agricultural sites have become first order tasks mainly for security or the safety reasons. The main issues of this activity is tied to the size of the sites and to their accessibility. Thus, it seems nowadays relevant to tackle with this problem with robots, which can detect potential issues with a low operational cost. To that purpose, in addition to individual patrolling behavior, robots need coordination schemes. The goal of this paper is to explore the possibility of using interference fringes and waves properties in a virtual environment to tackle with the longitudinal distance regulation in the platoon control issue. The proposed model, based on a multi-agent paradigm, is considering each vehicle as an agent wave generator in the virtual environment.

International audience; Development of intelligent behaviors for vehicles has to cope with rigorous specifications. Many tests are performed for verification, validation and a detailed study of resulting behaviors. In this workflow, testing algorithms with real vehicles is a cornerstone step in developing new intelligent features for future transportation systems. This step requires a lot of experiments before efficiently set up embedded autonomous systems. However, the development of such tests is not linear and it highly depends on external variable factors (weather, hardware or software issues, etc.). To overcome these issues, it is possible to carry out a first series of validation under a simulation framework. In this paper, we describe a simulation platform called VIPS (Vehicles In Platoon Simulator), which provides a virtual environment designed to test and validate platoon control approaches, as close as possible to reality, before bringing them in the real vehicle system. Among the features provided by the simulator, one main contribution is the direct matching between perception and control of both real and virtual vehicles. Moreover, VIPS is also able to deal with perturbations, which are dynamically added to running scenarios so as to reproduce real world conditions. Simulation results in the context of platoon control systems show the benefits of the proposed simulation framework.

International audience; Since a decade multi-agents became a widespread solution to tackle different kinds of issues in various application fields. Among the two main trends in multi-agent approaches (cognitive vs. reactive), the reactive one is particularly interesting for applications that require both fast response time, adaptability and robustness. Reactive Multi-agent Systems rely on simple interaction schemes between entities that can be inspired whether by Biology or Physics. The goal of this paper is to explore the possibility of using interference fringes and waves properties from a multi-agent standpoint so as to tackle a real problem, longitudinal platoon regulation, already explored by different strategies including particle agents approaches or standard PID controller.

Nowadays, automotive equipments tend to offer more and more drinving assistance. Among these, car park assist systems are now widespread on both commercial and research levels. This paper proposes a car park assistance based on the use of the multi-agent paradigm. This approach based on the transformation of vehicle perception into a virtual environment for agents and on the evaluation of agency properties. This evaluation is then turned into a decision for vehicle guidance. The proposal is evaluated in simulation by comparing it with industrial solutions.

Recently, many research efforts are directed towards co-evolutionary algorithms. The present work aims at the assessment of Hierarchical Cooperative CoEvolution (HCCE) being properly formulated to address hierarchical problems where simple components having separate design objectives, are parts of other more complex ones. HCCE is able to highlight the specialties of each component and additionally enforce their successful integration in a composite structure. Here we present HCCE describing also the internal dynamics that provide its effectiveness in addressing difficult distributed design problems. Additionally, the results described in the present work attest to its validity and superior performance against ordinary Unimodal evolution, and Enforced SubPopulation coevolution.

Even if the multi-agent paradigm has been evolving for fifteen years, the development of concrete methods for problem solving remains a major challenge. This paper focuses on reactive multi-agent systems because they provide interesting properties such as adaptability and robustness. In particular, the role of the environment, which is effectively where the system computes and communicates, is studied. From this analysis a methodology to design or engineer reactive systems is introduced. Our approach is based on the representation of the problem's constraints considered as perturbations to stabilize. Agents are then defined, in the second place, as a means of regulating the perturbations. Finally, the relevancy of our proposition is justified through the development of two solving models applied to real and complex problems.

This paper presents a local control approach to linear platoons, systems of vehicles that displace together adopting a train configuration without any material coupling. Linear platoon technology is considered as a potential base for the definition of new urban transportation services. The main problems related to platoon systems is the geometry control: control of inter-vehicle distance and common trajectory matching. The geometry control problem is generally treated according to one of two approaches: global or local. This paper focuses on a local approach which does not require sophisticated sensors and/or costly road equipment. The objective of this local control approach allows satisfactory distance regulation and matching to any common trajectory, by using local perception capabilities of platoon's vehicles. The basic perception capability consists in measuring the vectorial distance between a given vehicle and its predecessor. The behaviour of each platoon vehicle is determined by a physics inspired interaction model based on a virtual spring-damper. Stability and other platoon safety properties are analysed on the base of the physics inspired interaction model. Both simulation and experimentation have been used to measure trajectory error and to evaluate inter-vehicle distance during platoon evolution.

This work presents an approach to the obstacle avoidance problem, applicable in the frame of driver assistance. A decision, expressed as a proposed acceleration vector for the vehicle, is elaborated from the evaluation of a set of indicators characterizing the global state of a system of reactive agents (RMAS). Those agents evolve in a virtual environment produced on the base of vehicle’s perceptions of the material environment around it. Agent-to-agent and agent-to-environment interactions are defined in order to produce a distribution of agents over the virtual environment. This distribution, taken as the global state of the system, is analyzed by applying a set of indicators inspired from statistical physics, to calculate a new vehicle’s acceleration vector. This work presents the details of the RMAS model and its interaction laws, together with the global state evaluation functions. The approach has been applied to experimentation with a laboratory vehicle. Some experimental results are presented here.