Author: Michael Aeberhard, Apex.AI
Apex.AI has its roots in the automotive and robotics industries. Co-founders Jan and Dejan and many Apex.AI employees have extensive experience in autonomous driving and production ADAS within the automotive industry. Our mission is to improve the status quo of software development in automotive with the advent of the software-defined vehicle (SDV). Apex.AI’s co-founders have also been early adaptors, contributors and promoters of the Robot Operating System (ROS), which started as a common open-source framework and middleware for robotics but over the last decade and a half has expanded into the research and pre-production departments of many companies in a wide range of industries.
Apex.AI has established itself as a key provider of a “mobility OS” for various industries. The heavy machinery industry is particularly interesting, as modern technology and automation are starting to take over the machines that feed, build and provide for our society. Use cases in agriculture, mining, warehouses, production plants, construction, etc., will be increasingly automated, where sensor technologies such as cameras, lidars, radars and high-performance computing will be integrated into heavy mobile machines to enable fully autonomous operations of industry processes. These industries have a unique opportunity to introduce autonomy in a shorter time frame due to the economies involved (the machines and persons required to operate them have higher associated costs) and the limited operational domain (e.g., the machines operate in known, closed-off areas, usually at low speeds, where the work is being done with trained personnel in the surrounding areas). An underlying software framework, middleware and “OS” will be required for these machines in the near future, where Apex.AI’s products are a perfect match.
Since 2023, Apex.AI has been an active partner in the joint project Professional Operating Vehicles OS, or POV.OS, consisting of 16 partners from industry and academia.
POV.OS is a research project funded by the German Federal Ministry of Economic Affairs and Climate Action as part of the ministry’s research program on “New vehicles and system technologies.” The project develops an automation platform with a common hardware and software architecture for use in mobile heavy machinery. Demonstrators with autonomy functionality in agriculture, warehouse transportation and mining are being jointly developed by the project partners. A key aspect of the project is the choice of ROS 2 as the software development framework for implementing the autonomous functionality. Project partners are implementing and integrating the autonomous functionality as ROS 2 nodes and are running ROS 2 systems on the project’s vehicle demonstrators. Apex.AI plays a key role as the project’s “ROS experts,” supporting project partners with ROS 2 implementations, navigating the ROS community and ensuring ROS 2 meets the requirements for heavy industry machines.
The POV.OS joint project is organized into several work packages (WP) that coordinate the work in the project.
Apex.AI is directly involved in the work packages
WP 1 Common platform topics (Plattformübergreifende Themen)
WP 2 Representative use-cases (Repräsentative Anwendungsfälle)
WP 3 Safety of hardware and software (Sicherheit von Hardware und Software)
WP 4 Open and central computing architecture (Offene und zentralisierte Rechnerarchitektur)
WP 8 ROS 2 and autonomous functionality (ROS 2 und autonome Funktionalitäten)
Our goals in the POV.OS project are two-fold: 1) bring our expertise in ROS 2 and software-defined vehicles to the project consortium, and 2) learn about the specific requirements and use cases for mobile heavy machinery.
As part of WP 1, Apex.AI is working with project partners to define common interfaces between components of an autonomous system. Specifically for industry machines, standards such as J1939, ISOBUS and NMEA 2000 are important. Apex.AI dove deep into these standards and is applying them to autonomous systems implemented with ROS 2. We have concluded the initial research and are now working on an architecture for integrating these standards with a ROS 2-based system.
In WP 2, Apex.AI is closely involved with the autonomous systems and vehicle demonstrators being developed by the project partners.
WP 3 deals with the safety of software and hardware. Apex.AI has relevant safety and security expertise in the standards required by autonomous systems for mobile heavy machinery. The joint partners in WP 3 have completed thorough research of relevant standards for safety and security. Apex.AI, specifically, has completed a deep dive into ISO 25119 Tractors and machinery for agriculture and forestry, which is most relevant for agriculture use cases. The POV.OS project will put most of its emphasis on IEC 61508, as this is the base standard for most other industry-specific standards, including ISO 26262, ISO 25119, etc.
In WP 4, we are contributing to defining a general hardware and software architecture for all project partners and their specific use cases. An initial list of requirements for the software framework and operating system has been completed. Apex.AI has deep expertise in hardware-accelerated computing as part of the architecture and how it is integrated with ROS 2. We have chosen to work with the NXP S32G hardware for its work within the POV.OS project.
Apex.AI is also contributing its ROS 2 expertise in WP 8, guiding project partners on how to use ROS 2 to implement complex, autonomous systems. The POV.OS project decided to work with the Eclipse Cyclone DDS RMW with integrated Eclipse iceoryx™ zero-copy middleware, both of which Apex.AI has contributed to in the respective open-source projects. Apex.AI is taking the lessons learned from this project and is actively contributing to improving ROS 2 in the open source as part of the joint project.
Apex.AI took the lead last year in coordinating a joint POV.OS presentation at the first ROSCon DE, which was held in Karlsruhe on November 23-24, 2023. In the talk titled ROS 2 in POV.OS - Automatisierungsplattform für mobile Arbeitsmaschinen (ROS 2 in POV.OS - Automation platform for mobile heavy machinery), Apex.AI and the project partners presented the POV.OS to the German ROS community. Apex.AI’s Staff Application Engineer, Frederik Beaujean, participated in the talk (pictured on the left in the group photo with POV.OS project partners below).
Just a few weeks ago, the POV.OS project had its first consortium meeting in 2024 hosted by the Institute for Factory Automation and Production Systems (FAPS) of the Friedrich-Alexander-Universität Erlangen-Nürnberg in Nürnberg on April 22-23, 2024. The project partners spent two days reflecting on the POV.OS project progress thus far, participating in workshops, taking a tour of the FAPS laboratories, and having a traditional Nürnberg dinner at the Bratwurst Röslein.
The results of the POV.OS projects enable Apex.AI to understand the application of autonomy in industries deploying heavy machinery. This research with ROS 2 will lay the foundation for Apex.AI to extend its key products Apex.Grace and Apex.Ida to meet the requirements of these industries. The need for a framework in these industries is increasing, and heavy machines are becoming more automated and the companies behind these machines are actively looking for software solutions that can take them to production deployments in just a few years. Apex.AI products enable companies implementing autonomous heavy machinery to focus on their domain-specific application of autonomous systems in the field, instead of dealing with the development of a production-ready software development framework and middleware. We have already made strides in the agriculture industry by working with AGCO and Krone & Lemken to integrate Apex.Grace and Apex.Ida in their products.
With the insights and achievements from the POV.OS project, Apex.AI is poised to significantly broaden its presence and impact across multiple industries, ensuring that companies in sectors like agriculture, construction, and beyond can seamlessly integrate advanced autonomous solutions into their operations.
If you are interested in Apex.AI products for your projects, contact us. We’re always looking for talented people to join our international team. Visit our careers page to view our open positions.