Data Transfer Optimization Scheme Based on dSpace High-Fidelity LiDAR Model

Validating intelligent algorithms within the realm of Autonomous driving Domain Controllers (ADC) using virtual simulation platforms has become a foundational industry paradigm. Modeling the perception system of autonomous driving on a simulation platform, especially sensor modeling, has become a crucial aspect for validating intelligent algorithms. To accurately simulate real sensors, it is imperative to construct sensor models reasonably and ensure accurate data transmission between the sensor model and the ADC. This paper, based on the high-fidelity dSpace simulation platform, employs a LiDAR model to simulate a real LiDAR within the perception system. Post-processing operations, such as time synchronization and data validation, are applied to the LiDAR model to ensure the accuracy of the sensor model's data link. Addressing issues of mismatched data transmission frequencies and inconsistent data packet quantities between the LiDAR model and the ADC, we propose a solution named DCD-FIFO based on a First-In-First-Out (FIFO) queue. In this solution, the LiDAR model actively manages frame dropping, complements missing data packets, and delays sending data packets to the ADC to resolve data transmission issues. Experimental results demonstrate that the high-fidelity LiDAR simulation model in the dSpace platform accurately simulate a real LiDAR. Moreover, using the proposed DCD-FIFO solution, reliable data transmission is achieved between the LiDAR model and the ADC.