Autonomous cyber-physical systems depend on high-resolution sensors that generate multi-gigabyte-per-second data streams for accurate environmental perception and system safety. However, existing publish-subscribe middleware frameworks, such as ROS 2 with DDS, are optimized for small data objects and face significant latency and overhead challenges when handling large sensor data between distributed application nodes. Recognizing the need for more efficient data management, we propose a new companion middleware that prioritizes application-specific data relevance, enabling selective communication of critical information while reducing the burden on communication resources and maintaining interoperability with state-of-the-art publish-subscribe middleware. This companion middleware enables timely and effective sharing of sensor data by focusing on regions of interest relevant to specific tasks, such as traffic light detection in driving scenarios. Experimental evaluations of our open source implementation of this companion middleware on a Linux platform demonstrate that our protocol integrates efficiently with ROS 2, significantly enhancing data management and communication efficiency.

In the foreseeable future, highly automated mobile systems, such as vehicles, robots, UAVs, or trains, will be confronted with difficult situations that require external support. The availability of such external support corresponds to level 4 driving automation and is an essential feature in current robotaxis and automated public transportation. While the first generation of level 4 prototypes relied on safety driver support, commercial systems are gradually moving towards support by teleoperation. Designing teleoperation support for level 4 systems is an end-to-end problem involving two main research and practical challenges, the teleoperation function defining the remote human interface with its scene representation and available control functions, and the real-time communication channel involving wired and wireless segments, which must provide reliable end-to-end data transport. Both challenges are tightly linked, and combined solutions are needed to reach the required safe teleoperation. Solutions can make use of the rich sensing and control system of a level 4 vehicle, which however can only be exploited if the communication channel provides adequate real-time access.

Driving automation has become a major cost factor in automotive design. High computation demand for machine learning (ML) applications and growing sensor resolution and data rate require expensive hardware technology and networking. Newer network technologies and topologies can at best compensate the growing communication demand, but the network still accounts for a complex wiring harness with many dedicated sensor cables. In this paper, we exploit the context specific sensor data access of ML perception applications to minimize the sensor data traffic. For that purpose, we extend the popular Data Distribution Service (DDS) publish-subscribe middleware by a subscriber-centric software caching feature, which is then supported by an appropriate network scheduling. Using realistic data sets and ML applications from the popular Autoware benchmark and a zonal architecture according to the P802.1DG automotive Time-Sensitive Networking (TSN) network profile, we demonstrate that both cabling structure and network latency can be significantly reduced for both 1 Gbps and 10 Gbps TSN technologies. This result enables faster perception and/or lower cable cost, at no loss in data quality or reliability.

Vehicle-to-everything (V2X) roadmaps envision future applications that require the reliable exchange of large sensor data over a wireless network in real time. Applications include sensor fusion for cooperative perception or remote vehicle control that are subject to stringent real-time and safety constraints. Real-time requirements result from end-to-end latency constraints, while reliability refers to the quest for loss-free sensor data transfer to reach maximum application quality. In wireless networks, both requirements are in conflict, because of the need for error correction. Notably, the established video coding standards are not suitable for this task, as demonstrated in experiments. This article shows that middleware-based backward error correction (BEC) in combination with application controlled selective data transmission is far more effective for this purpose. The mechanisms proposed in this article use application and context knowledge to dynamically adapt the data object volume at high error rates at sustained application resilience. We evaluate popular camera datasets and perception pipelines from the automotive domain and apply two complementary strategies. The results and comparisons show that this approach has great benefits, far beyond the state of the art. It also shows that there is no single strategy that outperforms the other in all use cases.

With the current trend towards large data volumes, state-of-the-art communication and data management solutions based on the publish/subscribe pattern reach the network limits, hampering the deployment of advanced edge and cloud computing services. A main reason is the established publisher-centric distribution mechanism that over-utilizes network resources. We suggest a network and subscriber-centric adaptive data caching communication scheme supported by dynamic network management and configuration, which has great potential to solve the data distribution challenges. We show that state-of-the-art publish/subscribe middlewares like Data Distribution Service (DDS) can transparently be combined with such software caching.

This paper summarizes the talks of a special session on the IPF 2.0 project, a collaborative German-US research project that leverages self-awareness principles for the self-management of distributed systems of autonomous multiprocessor systems-on-chip (MPSoCs).

The high complexity of a network stack as found in Linux leads to unpredictable timing behavior and interference that is incompatible to safety standards. Inspired by the microkernel approach, a filter stack separates critical from non-critical traffic using a heterogeneous architecture. Such separation likely leads to performance inefficiencies or limitations in functionality. In this paper we improve a filter architecture by hardware supported temporal and spatial isolation. We show that this approach provides near optimum performance even when applied to a commercial Ethernet interface. Hardware support simplifies the stack architecture, guaranteeing high predictability for critical traffic.

Deterministic communication is a key challenge for modern embedded real-time systems. This is especially true for the automotive domain, where safety-critical functions are distributed over multiple electronic control units (ECUs) across the vehicle. To handle the increased complexity together with the higher data volume, Ethernet is considered as a universal media. Consequently, a great effort has been spent on making Ethernet predictable to satisfy timing and safety requirements. However, the communication stacks (COM-stacks) that build the bridge between applications and the network did not receive the same attention. Until now it has been dealt with as legacy software and overloaded with additional features to support the new multi-level protocols. Moreover, with the shift to automated vehicles, communication requirements will change fundamentally, transforming the car to a data centric system. In this paper we highlight the communication requirements of future autonomous cars and show why today's COM-stack designs are not suited to meet them. We present a novel design approach that ensures predictable timing and resource sharing while focusing on a minimalist and modular design. We also show how it fits to a state-of-the-art middleware, while keeping the complexity as small as possible.