Thermal error significantly impacts the machining precision of machine-tools. Thermal deformations in the machine-tool structure caused by the various machine heat sources is at the origin of this phenomenon. In order to ensure the expected quality of the parts, manufacturer have to run the machine-tools for hours before start producing in order to reach the machine thermal stability. This heating phase has a high negative impact on the machine productivity on one hand and on its ecological footprint on the other. This paper presents a data-driven approach to model and predict the thermal error in order to correct the tool reference position accordingly. The automatic adjustment of tool position allows to produce parts with the expected quality and precision regardless of the thermal state of the machines, which substantially increase their productivity. For this purpose, temperature sensors as well as high precision tool position measurement instruments are deployed on a Tornos SwissNano4 machine-tool. A set of experiments are conducted to collect data related to these two measurements. Four major Machine Learning algorithms are trained using a subset of the collected data and tested with the remaining data subset. Quantitative and comparative analysis shows that three of the four algorithms have a prediction with a mean Absolute Error (MAE) below 1µm and a Correlation Coefficient higher than 90%. Even classical linear regression models are able to predict the thermal error with high accuracy. Advanced Machine Learning techniques show high potential to provide a better prediction accuracy.

Next generation of embedded Information and Communication Technology (ICT) systems are interconnected and collaborative systems able to perform autonomous tasks. The remarkable expansion of the embedded ICT market, together with the rise and breakthroughs of Artificial Intelligence (AI), have put the focus on the Edge as it stands as one of the keys for the next technological revolution: the seamless integration of AI in our daily life. However, training and deployment of custom AI solutions on embedded devices require a fine-grained integration of data, algorithms, and tools to achieve high accuracy and overcome functional and non-functional requirements. Such integration requires a high level of expertise that becomes a real bottleneck for small and medium enterprises wanting to deploy AI solutions on the Edge , which, ultimately, slows down the adoption of AI on applications in our daily life. In this work, we present a modular AI pipeline as an integrating framework to bring data, algorithms, and deployment tools together. By removing the integration barriers and lowering the required expertise, we can interconnect the different stages of particular tools and provide a modular end-to-end development of AI products for embedded devices. Our AI pipeline consists of four modular main steps: (i) data ingestion, (ii) model training, (iii) deployment optimization, and (iv) the IoT hub integration. To show the effectiveness of our pipeline, we provide examples of different AI applications during each of the steps. Besides, we integrate our deployment framework, Low-Power Deep Neural Network (LPDNN), into the AI pipeline and present its lightweight architecture and deployment capabilities for embedded devices. Finally, we demonstrate the results of the AI pipeline by showing the deployment of several AI applications such as keyword spotting, image classification, and object detection on a set of well-known embedded platforms, where LPDNN consistently outperforms all other popular deployment frameworks.

Next generation of embedded Information and Communication Technology (ICT) systems are interconnected and collaborative systems able to perform autonomous tasks. The remarkable expansion of the embedded ICT market, together with the rise and breakthroughs of Artificial Intelligence (AI), have put the focus on the Edge as it stands as one of the keys for the next technological revolution: the seamless integration of AI in our daily life. However, training and deployment of custom AI solutions on embedded devices require a fine-grained integration of data, algorithms, and tools to achieve high accuracy and overcome functional and non-functional requirements. Such integration requires a high level of expertise that becomes a real bottleneck for small and medium enterprises wanting to deploy AI solutions on the Edge , which, ultimately, slows down the adoption of AI on applications in our daily life. In this work, we present a modular AI pipeline as an integrating framework to bring data, algorithms, and deployment tools together. By removing the integration barriers and lowering the required expertise, we can interconnect the different stages of particular tools and provide a modular end-to-end development of AI products for embedded devices. Our AI pipeline consists of four modular main steps: (i) data ingestion, (ii) model training, (iii) deployment optimization, and (iv) the IoT hub integration. To show the effectiveness of our pipeline, we provide examples of different AI applications during each of the steps. Besides, we integrate our deployment framework, Low-Power Deep Neural Network (LPDNN), into the AI pipeline and present its lightweight architecture and deployment capabilities for embedded devices. Finally, we demonstrate the results of the AI pipeline by showing the deployment of several AI applications such as keyword spotting, image classification, and object detection on a set of well-known embedded platforms, where LPDNN consistently outperforms all other popular deployment frameworks.

Bonseyes is an Artificial Intelligence (AI) platform composed of a Data Marketplace, a Deep Learning Toolbox, and Developer Reference Platforms with the aim of facilitating tech and non-tech companies a rapid adoption of AI as an enabler for their business. Bonseyes provides methods and tools to speed up the development and deployment of AI solutions on low power Internet of Things (IoT) devices, embedded computing systems, and data centre servers. In this work, we address the deployment and the integration of Bonseyes AI applications in a wider enterprise application landscape involving different applications and services. We leverage the well-established IoT platform FIWARE to integrate the Bonseyes AI applications into an enterprise ecosystem. This paper presents two AI application deployment and integration scenarios using FIWARE. The first scenario addresses use cases where edge devices have enough compute power to run the AI applications and there is only need to transmit the results to the enterprise ecosystem. The second scenario copes with use cases where an edge device may delegate most of the computation to an external/cloud server. Further, we employ FIWARE IoT Agent generic enabler to manage all edge devices related to Bonseyes AI applications. Both scenarios have been validated on concrete use cases and demonstrators.

Musculo Skeletal Disorders (MSDs) is the most common disease in the workplaces causing disabilities and excessive costs to industries, particularly in EU countries. Most of MSDs prevention programs have focused on a combination of interventions including training to change individual behaviors (such as awkward postures). However, little evidence proves that current training approach on awkward postures is efficient and can significantly reduce MSDs symptoms. Therefore, dealing with awkward postures and repetitive tasks is the real challenge for practitioners and manufacturers, knowing that the amount of risk exposure varies increasingly among workers depending on their attitude and expertise as well as on their strategy to perform the task. The progress in MSDs prevention might come through developing new tools that inform workers more efficiently on their gestures and postures. This paper proposes a potential Serious Game that immerses industrial workers using Virtual Reality and helps them recognize their strategy while performing tasks and trains them to find the most efficient and least risky tactics.

Littering quantification is an important step for improving cleanliness of cities. When human interpretation is too cumbersome or in some cases impossible, an objective index of cleanliness could reduce the littering by awareness actions. In this paper, we present a fully automated computer vision application for littering quantification based on images taken from the streets and sidewalks. We have employed a deep learning based framework to localize and classify different types of wastes. Since there was no waste dataset available, we built our acquisition system mounted on a vehicle. Collected images containing different types of wastes. These images are then annotated for training and benchmarking the developed system. Our results on real case scenarios show accurate detection of littering on variant backgrounds.

Due to real world physical constraints (e.g. walls), experimenting a virtual reality phenomenon implies transitional issues from one virtual environment (VE) to another. This paper proposes an experiment which studies the relevance of smooth and imperceptible transitions from a familiar and pleasurable virtual environment to a similar workplace as a mean to avoid traumatic experiences in VR for trainees. Specifically, the hereby work assumes that the user consciousness regarding virtual environment transitions is a relevant indicator of positive user experience during those. Furthermore, serious games taking place in purely virtual environments have the advantage of coping with various workplace configurations and tasks that the trainee can practice. However, the virtual world of serious games should be carefully designed in order to avoid traumatic experiences for trainees. The results presented stem from an empirical evaluation of user experience conducted with 80 volunteers. This evaluation shows that more than one-third of the participants did not even notice the VE global change.

An instrument to improve the quality of life in large cities, helping to reduce the car traffic, is presented in this paper. It will result in a mobile guidance software that will help the drivers looking for a parking place to find it efficiently. SmartPark relies on available parking information systems, as well as on new sensors or even on social data inputs. A fixed magnetic on-street sensor and a video processing smart camera have been developed and prototypes of both devices were tested. Their data is available through a cloud-based Internet of Things infrastructure and continuously updated every few seconds. Databases will be built over time enabling data mining methods to infer parking availability models over time which will be used, eventually, by the algorithms feeding the mobile application.

This paper presents a real-world proven solution for dynamic street light control and management which relies on an open and flexible Internet of Things architecture. Substantial contribution is brought at the interoperability level using novel device connection concept based on model-driven communication agents to speed up the integration of sensors and actuators to Internet of Things platforms. The paper shows also results from real-world tests with deployed dynamic street lights in urban spaces. The proposed dynamic light control solution permits an energy saving of about 56% compared to classical static, time-based street light control.

Internet of Things (IoT) seems a viable way to enable the Smart Cities of the future. iNUIT (Internet of Things for Urban Innovation) is a multi-year research program that aims to create an ecosystem that exploits the variety of data coming from multiple sensors and connected objects installed on the scale of a city, in order to meet specific needs in terms of development of new services (physical security, resource management, etc.). Among the multiple research activities within iNUIT, we present two projects: SmartCrowd and OpEc. SmartCrowd aims at monitoring the crowd’s movement during large events. It focuses on real-time tracking using sensors available in smartphones and on the use of a crowd simulator to detect possible dangerous scenarios. A proof-of-concept of the application has been tested at the Paléo Festival (Switzerland) showing the feasibility of the approach. OpEc (Optimisation de l’Eclairage public) aims at using IoT to implement dynamic street light management and control with the goal of reducing street light energy consumption while guaranteeing the same level of security of traditional illumination. The system has been tested during two months in a street in St-Imier (Switzerland) without interruption, validating its stability and resulting in an overall energy saving of about 56%.

This paper presents a novel approach towards dynamic street light control, which combines advanced Information and Communication Technologies (ICT) and citizens’ involvement and engagement. Our proposal is based on the Citizens’ involvement which would strongly increases the efficiency and performance of technological solutions in smart city context. We believe that Serious Games have the potential to strengthen people motivation in this context.

There exists, today, a wide consensus that Internet of Things (IoT) is creating a wide range of business opportunities for various industries and sectors like Manufacturing, Healthcare, Public infrastructure management, Telecommunications and many others. On the other hand, the technological evolution of IoT facing serious challenges. The fragmentation in terms of communication protocols and data formats at device level is one of these challenges. Vendor specific application architectures, proprietary communication protocols and lack of IoT standards are some reasons behind the IoT fragmentation. In this paper we propose a software enabled framework to address the fragmentation challenge. The framework is based on flexible communication agents that are deployed on a gateway and can be adapted to various devices communicating different data formats using different communication protocol. The communication agent is automatically generated based on specifications and automatically deployed on the Gateway in order to connect the devices to a central platform where data are consolidated and exposed via REST APIs to third party services. Security and scalability aspects are also addressed in this work.

In visual-based robot navigation, panoramic vision emerges as a very attractive candidate for solving the localization task. Unfortunately, current systems rely on specific feature selection processes that do not cover the requirements of general purpose robots. In order to fulfil new requirements of robot versatility and robustness to environmental changes, we propose in this paper to perform the feature selection of a panoramic vision system by means of the saliency-based model of visual attention, a model known for its universality. The first part of the paper describes a localization system combining panoramic vision and visual attention. The second part presents a series of indoor localization experiments using panoramic vision and attention guided feature detection. The results show the feasibility of the approach and illustrate some of its capabilities

Saliency-based visual attention models provide visual saliency by combining the conspicuity maps relative to various visual cues. Because the cues are of different nature, the maps to be combined show distinct dynamic ranges and a normalization scheme is therefore required. The normalization scheme used traditionally is an instantaneous peakto- peak normalization. It appears however that this scheme performs poorly in cases where the relative contribution of the cues varies significantly, for instance when the kind of scene changes, like when the scene under study becomes unsaturated or worse, when it looses any chromaticity. To remedy this drawback, this paper proposes an alternative normalization scheme that scales each conspicuity map with respect to a long-term estimate of its maximum, a value which is learned initially from a large number of images. The advantage of the new method is first illustrated by several examples where both normalization schemes are compared. Then, the paper presents the results of an evaluation where the computed visual saliency of a set of 40 images is compared to the respective human attention as derived from the eye movements by a population of 20 subjects. The better performance of the new normalization scheme demonstrates its capability to deal with scenes of varying type, where cue contributions vary a lot. The proposed scheme seems thus preferable in any general purpose model of visual attention.

In the heart of the computer model of visual attention, an interest or saliency map is derived from an input image in a process that encompasses several data combination steps. While several combination strategies are possible and the choice of a method influences the final saliency substantially, there is a real need for a performance comparison for the purpose of model improvement. This paper presents contributing work in which model performances are measured by comparing saliency maps with human eye fixations. Four combination methods are compared in experiments involving the viewing of 40 images by 20 observers. Similarity is evaluated qualitatively by visual tests and quantitatively by use of a similarity score. With similarity scores lying 100% higher, non-linear combinations outperform linear methods. The comparison with human vision thus shows the superiority of non-linear over linear combination schemes and speaks for their preferred use in computer models.

Visual attention is the ability of a vision system, be it biological or artificial, to rapidly detect potentially relevant parts of a visual scene. The saliency-based model of visual attention is widely used to simulate this visual mechanism on computers. Though biologically inspired, this model has been only partially assessed in comparison with human behavior. The research described in this paper aims at assessing its performance in the case of natural scenes, i.e. real 3D color scenes. The evaluation is based on the comparison of computer saliency maps with human visual attention derived from fixation patterns while subjects are looking at the scenes. The paper presents a number of experiments involving natural scenes and computer models differing by their capacity to deal with color and depth. The results point on the large range of scene specific performance variations and provide typical quantitative performance values for models of different complexity.

This paper reports a landmark-based localization method relying on visual attention. In a learning phase, the multicue, multi-scale saliency-based model of visual attention is used to automatically acquire robust visual landmarks that are integrated into a topological map of the navigation environment. During navigation, the same visual attention model detects the most salient visual features that are then matched to the learned landmarks. The matching result yields a probabilistic measure of the current location of the robot. Further, this measure is integrated into a more general Markov localization framework in order to take into account the structural constraints of the navigation environment, which significantly enhances the localization results. Some experiments carried out with real training and test image sequences taken by a robot in a lab environment show the potential of the proposed method.

Visual attention refers to the ability of a vision system to rapidly detect visually salient locations in a given scene. On the other hand, the selection of robust visual landmarks of an environment represents a cornerstone of reliable vision-based robot navigation systems. Indeed, can salient scene locations provided by visual attention be useful for robot navigation? This work investigates the potential and effectiveness of the visual attention mechanism to provide pre-attentive scene information to a robot navigation system. The basic idea is to detect and track the salient locations, or spots of attention by building trajectories that memorize the spatial and temporal evolution of these spots. Then, a persistency test, which is based on the examination of the lengths of built trajectories, allows the selection of good environment landmarks. The selected landmarks can be used for feature-based localization and mapping systems which helps mobile robot to accomplish navigation tasks.

Visual attention is the ability of a vision system, be it biological or artificial, to rapidly detect potentially relevant parts of a visual scene, on which higher level vision tasks, such as object recognition, can focus. The saliency-based model of visual attention represents one of the main attempts to simulate this visual mechanism on computers. Though biologically inspired, this model has only been partially assessed in comparison with human behavior. Our methodology consists in comparing the computational saliency map with human eye movement patterns. This paper presents an in-depth analysis of the model by assessing the contribution of different cues to visual attention. It reports the results of a quantitative comparison of human visual attention derived from fixation patterns with visual attention as modeled by different versions of the computer model. More specifically, a one-cue gray-level model is compared to a two-cues color model. The experiments conducted with over 40 images of different nature and involving 20 human subjects assess the quantitative contribution of chromatic features in visual attention.