Parallel Code Generation for ROS 2 Nodes from Simulink Models
Analysis
This paper addresses the challenge of parallelizing code generation for complex embedded systems, particularly in autonomous driving, using Model-Based Development (MBD) and ROS 2. It tackles the limitations of manual parallelization and existing MBD approaches, especially in multi-input scenarios. The proposed framework categorizes Simulink models into event-driven and timer-driven types to enable targeted parallelization, ultimately improving execution time. The focus on ROS 2 integration and the evaluation results demonstrating performance improvements are key contributions.
Key Takeaways
- •Proposes an MBD framework for parallel code generation from Simulink models for ROS 2.
- •Categorizes ROS 2-compatible Simulink models into event-driven and timer-driven types for targeted parallelization.
- •Addresses the challenges of manual parallelization and limitations of existing MBD approaches in multi-input scenarios.
- •Demonstrates improved execution time through parallelization.
“The evaluation results show that after applying parallelization with the proposed framework, all patterns show a reduction in execution time, confirming the effectiveness of parallelization.”