Performance Analysis of Embedded Control Algorithms used in UAVs

Abstract

Performance analysis of embedded systems is critical when dealing with Cyber-Physical Systems that require stability guarantees. They typically operate having to respect deadlines imposed during the design of the related control system. In a recent past performance analysis was typically done only by executing the code, and making measures, on the target embedded platform. Nowadays, code execution/measuring can also be done on simulation software, which offers greater degree of liberty for designers to configure the system for the desired tests. This paper presents results obtained from analyzing the performance of two control algorithms developed for controlling an Unmanned Aerial Vehicle (UAV) running on simulated and real embedded platforms. Such analysis is important for twofold reasons: better understand the timing behavior or the algorithms and evaluate architectural issues related with the target embedded platform. Raspberry Pi 3 Model B+ (with Cortex-A53 processor) is used as reference platform and serves as basis for creating different simulated versions for the analysis. Initial results highlighted the important role played by cache memory in the performance of the control algorithms and were able to detect a major bottleneck in one of the control algorithms that could compromise system stability.