Rail transport is a crucial building in helping us conserve natural energy resources and achieve global climate targets. Driving by train or tram becomes even safer and more comfortable thanks to advanced driver assistance systems as we know them from the automotive domain. Future rolling stock will be equipped with sensors and broadband communication to allow for a much faster detection of potential hazards and prompt the driver or an automatic train control system to act in dangerous situations, which can be caused by reduced visibility due to weather conditions, and when approaching or leaving stations. Moreover, advanced in-cabin monitoring systems will look out for the driver and passenger´s physical well-being for and driving comfort.
Multisensor applications for driver assistance systems in future trains will typically connect to onboard sensors such as long-range radar and lidar sensors, and they will provide features such as broadband wireless communication with the traffic infrastructure and GNSS based positioning. Therefore, these applications will have to process asynchronous data streams in real time and in a time-correlated manner. For sensor fusion, which is a key step in building a reliable environment model, this sensor data will have to be synchronized. Specific cameras with infrared capabilities and artificial intelligence (AI) functions will be used to observe and assess the driver behavior.
The increasingly complex railway applications with a need for sensor integration, AI, and edge computing require field-tested development tools that let you focus on your actual tasks while taking over the rest, such as managing communication between system components.
With the RTMaps (Real-time Multisensor Applications) development and execution framework, you can integrate vehicle sensors, I/O, data, and communication buses in an easy way and use hundreds of off-the-shelf components. A user-friendly block-based approach lets you integrate your algorithms and connect with all necessary interfaces like cameras, radar, lidars, or GNSS in a few klicks for faster and more intuitive development. RTMaps handles asynchronous sensor data streams in a time-correlated manner and supports their synchronization when required. In addition to C/C++, the development framework natively supports the Python scripting language, which is widely used in the development of AI algorithms and lets you quickly integrate deep learning functions to cope with the challenging requirements on highly robust in-cabin monitoring functions. RTMaps works with multicore and multisensor systems to get the most out of your application that can be executed on different embedded platforms with Linux, Windows, or QNX. The robust AUTERA prototyping platform with outstanding performance is also supported. The modular platform provides numerous interfaces and configuration options and is designed for operatiion in harsh environments.
Drive innovation forward. Always on the pulse of technology development.
Subscribe to our expert knowledge. Learn from our successful project examples. Keep up to date on simulation and validation. Subscribe to/manage dSPACE direct and aerospace & defense now.