Differences

This shows you the differences between two versions of the page.

--- en:safeav:as:general [2025/10/17 08:35] – created agrisnik
+++ en:safeav:as:general [2025/10/17 08:57] (current) – [Middleware and Frameworks] agrisnik
@@ Line 1: / Line 1: @@
 ====== General Concepts of Architecture for Autonomous Systems ======
 Software architecture represents the high-level structure of a system, outlining the organisation of its components, their relationships, and the guiding principles governing their design and evolution. In autonomous systems, architecture defines how perception, planning, and control modules interact to achieve autonomy while maintaining safety, reliability, and performance ((Bass, L., Clements, P., & Kazman, R. (2021). Software Architecture in Practice (4th ed.). Addison-Wesley)).
@@ Line 15: / Line 14: @@
   - **Standardisation:** Adoption of middleware standards (e.g., ROS, DDS, MOOS) facilitates interoperability across platforms. Besides, the mentioned standardisation enables avoiding vendor locks and reduces overall costs due to increased competition.
   - **Data-centric communication:** Modern architectures rely on publish/subscribe paradigms to manage distributed communication efficiently.
+===== Middleware and Frameworks =====
+Middleware serves as the backbone that connects diverse modules, ensuring efficient data exchange and synchronisation. Prominent middleware systems in autonomous vehicles include:
+  * **ROS (Robot Operating System):** An open-source framework providing a modular structure for robotic applications, including perception, planning, and control ((Quigley, M., Gerkey, B., & Smart, W. D. (2009). Programming Robots with ROS: A Practical Introduction to the Robot Operating System. O’Reilly Media)).
+  * **DDS (Data Distribution Service):** A real-time communication standard widely used in aerospace and defence systems, supporting deterministic data exchange ((Object Management Group. (2023). Data Distribution Service (DDS) Standard. OMG)).
+  * **MOOS-IvP:** A marine-oriented autonomy framework designed for mission planning and vehicle coordination in autonomous underwater and surface vehicles ((Benjamin, M. R., Curcio, J. A., & Leonard, J. J. (2012). MOOS-IvP autonomy software for marine robots. Journal of Field Robotics, 29(6), 821–835)).
+  * A**UTOSAR Adaptive Platform:** A standard architecture for automotive systems emphasising safety, reliability, and scalability ((AUTOSAR Consortium. (2023). AUTOSAR Adaptive Platform Specification. AUTOSAR)).
+These middleware platforms not only promote interoperability but also enforce architectural patterns that ensure predictable performance across heterogeneous domains.
+Most autonomous systems follow a hierarchical layered architecture:
+^ Layer ^ Function ^ Examples ^
+| Hardware Abstraction | Interface with sensors, actuators, and low-level control | Sensor drivers, motor controllers |
+| Perception | Process raw sensor data into meaningful environment representations | Object detection, SLAM |
+| Decision-Making / Planning | Generate paths or actions based on goals and constraints | Path planning, behavior trees |
+| Control / Execution | Translate plans into commands for actuators | PID, MPC, low-level control loops |
+| Communication / Coordination | Handle data sharing between systems or fleets | Vehicle-to-vehicle (V2V), swarm coordination |
+Depending on functional tasks system’s architecture is split into multiple layers to abstract functionality and technical implementation as discussed above. Below is a schema of a generic architecture to get a better understanding of typical tasks at different layers.
+<figure Generic Autonomous System Architectures>
+{{ :en:safeav:as:rtu_ch1_figure1.png?400| Edge IoT system' architecture}}
+<caption>Generic Autonomous System Architecture </caption>
+</figure>
+===== The Role of AI and Machine Learning =====
+Modern autonomous systems increasingly integrate machine learning (ML) techniques for perception and decision-making. Deep neural networks enable real-time object detection, semantic segmentation, and trajectory prediction ((LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444)). However, these data-driven methods also introduce architectural challenges:
+  * Increased computational load requiring edge GPUs or dedicated AI accelerators.
+  * The need for robust validation and explainability to ensure safety.
+  * Integration with deterministic control modules in hybrid architectures.
+Thus, many systems adopt hybrid designs, combining traditional rule-based or dynamics-based control with data-driven inference modules, balancing interpretability and adaptability

en/safeav/as/general.1760690115.txt.gz · Last modified: 2025/10/17 08:35 by agrisnik