A new, old mobility concept for rural areas

Rural regions in particular suffer from the fact that public transport is not as well developed as in urban areas. As mobility has increasingly been moving towards private transport in the last decades, existing rail infrastructure in the countryside has been dismantled or lies fallow.

With the MONOCAB, this unused rail infrastructure can be revived for a new mobility concept. In Germany, the total length of disused but functional railway lines is over 5,000 km. MONOCABs can be operated simultaneously in both directions on single-track lines thanks to their very narrow vehicle design and the fact that they run on just one rail using gyro stabilization. This paves the way for an innovative rail-based mobility concept with service on demand, which represents an attractive alternative to conventional track reactivation. The use of MONOCABs, for example, can create a flexible, economically and ecologically efficient connection between rural areas and medium-sized urban areas. In addition, the low infrastructure and space requirements make MONOCABs an intriguing innovative, cost-effective urban and campus streetcar solution.

The first gyro-stabilized monorail vehicles were built as early as 1907 by Irish-Australian designer Louis Brennan. These vehicles were never used in regular operation, as the complex technology could not be safely controlled at the time. In the MonoCab OWL research project, the idea of a monorail vehicle for the individualization of rail-bound mobility is being re-evaluated under changed technical boundary conditions and mobility requirements. The project is being carried out by the Ostwestfalen-Lippe University of Applied Sciences, Bielefeld University of Applied Sciences, Fraunhofer IOSB-INA ,and Landeseisenbahn-Lippe e. V.

Model-Based Design Right from the Start

To turn the idea of the monorail vehicle into reality, the MonoCab OWL project designed, developed, and built two test vehicles within just two and a half years and put them into operation as part of a test and trial operation on a disused line. This required interdisciplinary collaboration within the 12-person research team from the fields of technology and design. To this end, a comprehensive model-based design process was established, which relies on consistent further development of the simulation models, starting from the requirements, concept, and predevelopment up to design and implementation, and ultimately to testing using HIL simulation.

The following Figure 1 shows the sequence of the development process, in particular the control software.

The team created important documents on system definition during the design phase at the beginning of the project:

• The system design defined not only the requirements for the test vehicles but also the modularized and hierarchized functional structure, including the interfaces for the flow of energy and information between the subsystems.
• The Modeling Guideline defines how the models must be set up in terms of their structure and interfaces in order to facilitate collaboration within the team and make sure the models can be used throughout all development steps.

At the beginning of the development process, models were used in particular for the system design of the stabilization components as well as the HV battery and traction drives. With the models’ help, the team selected system components for the test vehicles. Most of the models were built in MATLAB®/Simulink® and could be coupled with experimental test setups on the dSPACE real-time systems.

In the next step, the project team further developed these models to use them in the development and test of initial open- and closed-loop control concepts for driving and stabilizing the vehicles in the offline simulation. This included co-simulations between the Simulink environment and a multi-body simulation model in Simpack to analyze the driving behavior, taking into account the wheel-rail contact and the influence of frame stiffness on stabilization. Thanks to the consistent structuring and previously defined standardized interfaces, it was then possible to transfer the model components to a HIL simulation with minimal adjustments. The control components were implemented on dSPACE MicroAutoBox II systems, while a more modular real-time system from dSPACE was used as the HIL simulator. HIL simulation was used in particular to test the correct interaction of the four MicroAutoBox II systems as the core of the control unit network of a test vehicle. The interaction with other components, such as the remote control and a safety PLC, was also intensively checked before the vehicle was put into operation. The HIL tests are extremely important because they achieve the necessary depth of testing before the real test vehicles are put into test operation, thereby ensuring their safety. For this reason, the option of HIL testing in the vehicle was also provided after integrating the control unit network in the MONOCABs.

To visualize the vehicle behavior during the real-time simulation, a rail track was modeled in MotionDesk (Figure 1). In this way, the project team was able to use a clear representation of the vehicle with control systems running in real time for presentation purposes with manageable effort. The subsystems for energy supply as well as vertical and longitudinal dynamics were then commissioned on test benches.

The stabilization of the MONOCABs is based on a combination of two counter-rotating gyroscopes and a sliding mass in the vehicle floor. The vertical dynamics test bench (Figure 2) with its mechanical limitation of the vehicle angle φ served as a safe test environment for commissioning the complex stabilization control system. The basic structure of the test bench consisted largely of components identical to those later used in the vehicle and, thanks to dummy structures, had similar physical properties to the later vehicle. Model-based offline analyses were carried out iteratively and their results validated step by step in HIL operation on the test bench. At the same time, another part of the team developed a setup for commissioning the battery, the traction drives, and the charging technology.

Following the successful partial commissioning of the associated vehicle components, the control software developed in parallel was integrated and tested in the HIL simulation. The software was then put into operation and tested on the full scale test vehicles. This made it possible to achieve a very short assembly and commissioning sequence and to assemble individual components up to an operational test vehicle in less than seven months.

Comprehensive Development Process Requires Powerful Tools

The implementation of the described development process required hardware and software that the team could easily use across all development phases and that supported a swift implementation, commissioning, and testing of the electronic control unit (ECU) software. 

In this regard, a range of hardware and software products by dSPACE contributed significantly to the success of the project.

The total of eight MicroAutoBox II systems set up in the project were used as part of rapid control prototyping (RCP) to quickly implement the control unit functions. Initially, they were used in the test setups and in the HIL simulator to develop and validate the control functions. They were then integrated into the test vehicles as electronic control units (ECUs) to perform various tasks, as can be seen in Figure 3. Thanks to an interface on the front of the vehicle, the team was also able to use them for HIL simulations when installed. Thanks to the wide range of I/O options the MicroAutoBox II provides, it was also possible to easily integrate additional components at a later stage in the project. An industrial PC installed in the vehicles with the ControlDesk software made it easy to control and monitor the system at run time.

Modeling in MATLAB/Simulink with the RTI (Real-Time Interface) and other associated blocksets (e.g., RTI CAN Blockset and CAN MultiMessage Blockset) enabled simple implementation of the ECU software and the communication structure as well as flexible adaptation of the functions during testing. The integrated code generation and compilation enabled the team to quickly put the software into operation both on the test benches and in the test vehicles.

Outlook for the Future of the MONOCAB

The MonoCab OWL project developed two demonstrators that successfully prove the technical feasibility of the MONOCAB design. Further development steps must be taken in terms of validation, energy efficiency, and a higher degree of integration of all MONOCAB assemblies in order to develop the system into a product-oriented prototype. The field of automatic train operation (ATO) for the automated operation of MONOCABs in particular requires further research and development work. As part of other funding projects applied for (Figure 4), decisive key technologies for the safe implementation of MONOCAB technology will have to be researched in order to achieve a higher technological readiness level (TRL). In the development of basic ATO functionalities, the MONOCAB is used as an application example for driverless transport systems in the enableATO project of the German Center for Future Mobility (DZM). The MONOCAB ready project aims to create a new generation of test vehicles by 2026, which will enable the testing of ATO functionality as the basis for a new type of rail-bound on-demand mobility in the test field. 

From 2027, a regular operation trial with MONOCABs is to be prepared in parallel with an economic implementation. You can find more up-to-date information at www.monocab-owl.de

Funded by