Abstract

International audience; Message from the Program Chairs We are delighted to welcome you to the 27th edition of the International Conference on Real-Time Networks and Systems (RTNS), in Toulouse, France. RTNS is a friendly conference with a great sense of community that offers excellent opportunities for collaboration as well as a high-quality technical program. RTNS publishes papers addressing temporal issues in any area of computation and communication, at any level of abstraction, and in any application domain.This year we received 37 submissions and accepted 20 papers. The reviewing process involved 44 program committee members, who were assisted by 22 additional reviewers. Each submission received at least three reviews. Among the accepted papers, 4 outstanding papers were chosen, which are highlighted in the program, from which a separate committee selected one best paper and one best student paper. The conference program features a keynote talk giving a view on future challenges for the real-time community by Prof. Marko Bertogna (University of Modena, Italy), an industrial session on real-time challenges in four selected industrial application domains: automotive engine control systems (by Continental AG); automatic air transport (by Thales AVS France); electric smart grid substation automation systems (by SCLE-SFE, Groupe ENGIE); deterministic networking (by Cisco), and 6 technical sessions for accepted papers: analysis and validation; fault-tolerance, security, and data-flow; networking I and II; processor scheduling; and synchronization, preemption, and coordination. The main conference is also again complemented by the Junior Researcher Workshop on Real-Time Computing (JRWRTC), which is now in its 13th successful year.We would like to thank our many colleagues who have contributed to the success of RTNS 2019. First of all, we thank the Steering Committee for trusting us to manage the technical aspects of the conference. We also owe sincere thanks to all the PC members and reviewers who completed the crucial reviewing and shepherding work for the conference. The conference simply would not have been possible without the significant and sustained efforts of our General Chair, Jérôme Ermont (IRIT, ENSEEIHT, France), and the 5 members of the Local Organization Committee, who coordinated all aspects of the conference venue, web site, and the event itself. We also are grateful to the JRWRTC co-Chairs Antonio Paolillo (ULB, Belgium) and Benjamin Rouxel (UVA, Netherlands), and to the 15 members of the Workshop Committee for their efforts. Last but not least, we would like to thank the authors of all submitted papers – by entering your work into the process of review, critique, and possible rejection, even papers that were not accepted have helped to drive the important process of advancement in our field. We know it can be frustrating to have a paper rejected, but we hope that for each paper the reviewing feedback was helpful. Finally, we would like to thank the participants attending RTNS 2019 in Toulouse. We hope you will enjoy the technical program, the social events, and the conference venue and its surroundings! Ye-Qiong Song (LORIA - Université de Lorraine, France)Christopher Gill (Washington University in St. Louis, USA)RTNS 2019 Program Committee co-Chairs.

Abstract

International audience; Unmanned Aerial Vehicles (UAVs) are becoming an attractive solution for road traffic monitoring because of their mobility, low cost, and broad view range. Up to now, existing traffic monitoring systems based on UAVs only use one UAV with fixed trajectory to extract information about vehicles. In this paper, we propose a road traffic monitoring system using multiple UAVs. We develop a method to generate adaptive UAVs trajectories, which is based on the tracking of moving points in the UAV field of view. Also we generate UAVs trajectories using mobility models that are usually used to model vehicles mobility. UAVs monitor the traffic on a city road, they are responsible for collecting and sending, in real time, vehicle information to a traffic processing center for traffic regulation purposes. We show that the performance of our system is better than the performance of the fixed UAV trajectory traffic monitoring system in terms of coverage rates and events detection rates.

Abstract

International audience; Future autonomous vehicles and ADAS (Advanced Driver Assistance Systems) need real-time audio and video transmission together with control data traac (CDT). Audio/video stream delay analysis has been largely investigated in AVB (Audio Video Bridging) context, but not yet with the presence of the CDT in the new TSN context. In this paper we present a local delay analysis of AVB frames under hierarchical scheduling of credit-based shaping and time-aware shaping on TSN switches. We present the eeects of time aware shaping on AVB traac, how it changes the relative order of transmission of frames leading to bursts and worst case scenarios for lower priority streams. We also show that these bursts are upper-bounded by the Credit-Bases Shaper, hence the worst-case transmissions delay of a given stream is also upper-bounded. We present the analysis to compute the worst case delay for a frame, as well as the feasibility condition necessary for the analysis to be applied. Our methods (analysis and simulation) are applied to an automotive use case, which is deened within the Eurostars RETINA project, and where both control data traac and AVB traac must be guaranteed. CCS CONCEPTS • Computer systems organization →Embedded systems; • Networks →Network reliability;

Abstract

International audience; Integrity of the information provided by a perception system is crucial for advanced driver assistance systems intended for safety applications, like obstacle avoidance systems. A method to ensure integrity is to use different kinds of perception sources. Lidars are key sensors for multiple objects detection and tracking. Stereo vision systems (SVS) can be used to improve the tracking but, in this paper, we use also SVS to confirm the real existence of potential obstacles thanks to 3D dense reconstruction in focused regions of interest. Synchronization issues between the different sensors are addressed using predictive filtering. The proposed approach is evaluated in real conditions thanks to five use cases relevant to urban situations. Results show that this visual confirmation strategy is efficient.

Abstract

International audience; This SIG focuses on the engineering of automation in interactive critical systems. Automation has already been studied in a number of (sub-) disciplines and application fields: design, human factors, psychology, (software) engineering, aviation, health care, games. One distinguishing feature of the area we are focusing on is that in the field of interactive critical systems properties such as reliability, dependability, fault-tolerance are as important as usability, user experience or overall acceptance issues. The SIG targets at two problem areas: first the engineering of the user interaction with (partly-) autonomous systems: how to design, build and assess autonomous behavior, especially in cases where there is a need to represent on the user interface both autonomous and interactive objects. An example of such integration is the representation of an unmanned aerial vehicle (UAV) (where no direct interaction is possible), together with aircrafts (that have to be instructed by an air traffic controller to avoid the UAV). Second the design and engineering of user interaction in general for autonomous objects/systems (for example a cruise control in a car or an autopilot in an aircraft).

Abstract

International audience; Reliable obstacle detection and localization is a key issue for driver assistance systems, particularly in urban environments. In this study a multi-modal perception approach is investigated, the objective being to enhance vehicle localization and dynamic object tracking in a world-centric map. A 3D ego-localization is achieved by merging information from a stereo vision system and data obtained from vehicle sensors. Mobile objects are detected using a multi-layer lidar that is also used to identify a constrained search space within the multiple target tracking process. Object localization and tracking is then performed in the fixed frame, which facilitates analysis and understanding of the scene. Experimental results using real world data are performed to evaluate the performance of the multi-modal system, and these are presented to show the effectiveness of the approach.

Abstract

International audience; The use of over-the-air updates has attracted very much interest these last few years with the software-intensive development of embedded systems in the car industry. The development of autonomous driving and ADAS (Advanced Driver Assistance Systems) renders over-the-air updates mandatory, for both user satisfaction and economic reasons. How to make sure that remote updates of critical ADAS do not have an impact on safety? This is the question we tackle in our work with a major car manufacturer. This paper is a progress report. We summarize our approach involving AFT (Adaptive Fault Tolerance) implemented on ROS (Robot Operating System), describe the simulation platform we have developed to experiment and validate over-the-air updates of ADAS and AFT, and finally draw some lessons learnt and perspectives.

Abstract

National audience; This research paper proposes IEEE802.3az incorporated green algorithmic schemes on the software defined networking-based segment routing (SDN-based SR) centralized network. The proposed schemes are the green (i.e. the EAGER and CARE metrics) SDN-based SR networks with and without traffic engineering (TE) tunneling. Green metrics extend the IEEE 802.3az energy saving to the whole network, while tunneling avoids the load sharing mechanism included in the segment routing that is not convenient for energy saving purpose. The findings of emulation revealed that the energy efficiency under the EAGER-tunneling strategy is comparable to the conventional metric with TE tunneling, and the saving realized under the CARE-tunneling scheme is twofold in comparison with the LEGACY-tunneling strategy. Meanwhile, the energy saving performance achieved with all three metrics under TE non-tunneling was inferior to those with tunneling. Nonetheless, the lower end-to-end delay and lower maximum link utilization (MLU) were achievable with the non-tunneling strategy.

Abstract

International audience; ADAS (Advanced Driver Assistance Systems) algorithms increasingly use heavy image processing operations. To embed this type of algorithms, semiconductor companies offer many heterogeneous architectures. These SoCs (System on Chip) are composed of different processing units, with different capabilities, and often with massively parallel computing unit. Due to the complexity of these SoCs, predicting if a given algorithm can be executed in real time on a given architecture is not trivial. In fact it is not a simple task for automotive industry actors to choose the most suited heterogeneous SoC for a given application. Moreover, embedding complex algorithms on these systems remains a difficult task due to heterogeneity, it is not easy to decide how to allocate parts of a given algorithm on the different computing units of a given SoC. In order to help automotive industry in embedding algorithms on heterogeneous architectures, we propose a novel approach to predict performances of image processing algorithms applicable on different types of computing units. Our methodology is able to predict a more or less wide interval of execution time with a degree of confidence using only high level description of algorithms, and a few characteristics of computing units.

Abstract

Embedded systems have to support more and more demanding multimedia applications. Heterogeneous multi-core architectures are now commonplace in mobile electronic devices. The impact on the power and energy consumption is tremendous; it has to be evaluated as soon as possible in the design process. Multimedia development frameworks are used to abstract the complexity of the hardware to the designer. In this paper, we propose a methodology to develop high-level performance and consumption models for multimedia streaming applications based on the GStreamer framework. Our approach is based on measurements of the power consumptions and execution times during the processing of combined video and audio streams. Performance and consumption models are build for various plugins instantiated in the corresponding GStreamer pipelines. The combination of estimations for all those plugins leads to a precise evaluation of the complete plugin performances. The precision of the models is evaluated against measurements for different real-life streaming applications.