embedded AI has become a relevant research line with real applications involving the deployment of AI algorithms in units with limited computing resources, such as microcontrollers (MCUs). embedded AI has led to TinyM...
详细信息
ISBN:
(纸本)9783031820724;9783031820731
embedded AI has become a relevant research line with real applications involving the deployment of AI algorithms in units with limited computing resources, such as microcontrollers (MCUs). embedded AI has led to TinyML, a recently emerged paradigm that proposes to embed optimized ML models in MCUs. Traditionally, this deployment has required a deep knowledge of low-level programming, but currently some higher-level software libraries ease this deployment. However, analysis and practical examples of these libraries are still scarce. the main objective of this paper is to present the practical deployment of some of these libraries on a Proof of Concept and a test bench to analyze their effects on the inference latency and accuracy of the AI models. Results show that the development and deployment of these embedded models is already a feasible task requiring only basic AI and programming experience. Furthermore, the inference latency and accuracy of these models meets the requirements of several real-world applications.
the intelligent embedded system is a computing system combining artificial intelligence science and embedded technology, the system combines general-purpose processor and FPGA to achieve stronger processing capability...
详细信息
ISBN:
(纸本)9789819603534;9789819603541
the intelligent embedded system is a computing system combining artificial intelligence science and embedded technology, the system combines general-purpose processor and FPGA to achieve stronger processing capability, but it also brings system hardware and software design challenges. this paper proposes an intelligent embedded system task scheduling algorithm based on heterogeneous multi-core, which reasonably allocates hardware resources on the FPGA to optimize energy consumption under the premise of meeting system reliability requirements. the algorithm adopts a critical path-based energy consumption pre-allocation strategy, and the task scheduling makes the system schedule length shortest based on system reliability. Experiments show that the method of this paper outperforms other algorithms by an average of 5.34% in terms of energy consumption, and by an average of 6.83% in terms of scheduling length, which reflects the reasonableness and advancement of the method of this paper.
Large Language Models (LLMs) based Multi-Agent systems (MAS) are a rapidly emerging field with great potential to optimize work-flows across various industries. However, the unpredictable and hallucinating nature of L...
详细信息
Withthe rapid development of modern medical equipment technology, more and more diagnosis and treatment processes rely on the support of printing devices. When the doctor transfers data to the printer, the printer fi...
详细信息
TinyML is a novel area of machine learning that gained huge momentum in the last few years thanks to the ability to execute machine learning algorithms on tiny devices (such as Internet-of-things or embeddedsystems)....
详细信息
While the theory and practice of runtime monitoring are overall well developed, in embeddedsystems, runtime monitoring is not as common as one would expect. Especially small-scale embeddedsystems, which are found in...
详细信息
ISBN:
(纸本)9783031720437;9783031720444
While the theory and practice of runtime monitoring are overall well developed, in embeddedsystems, runtime monitoring is not as common as one would expect. Especially small-scale embeddedsystems, which are found in many household devices, are often somewhat safety-critical and could benefit from the possibility to detect software or hardware defects that cannot be uncovered with verification alone. While monitoring frameworks such as RTLola and Copilot are available to address this problem, employing them leads to a gap between verification and runtime monitoring by these frameworks having specialized specification languages for monitoring, and what can be expressed in them is incomparable to the capabilities of the temporal logics traditionally employed in formal verification. this paper discusses how (linear) temporal logic runtime monitoring for small-scale embeddedsystems can be made more efficient and attractive to the embeddedsystems practitioner. this includes identifying why monitoring for traditional temporal logics is somewhat inefficient in software, how this problem can be addressed in a low-cost way, and how runtime monitors can become a more useful component of an embedded system. By providing a way to translate a specification to monitor code that also tracks the reason for a specification violation, the overall approach discussed in the paper makes spending developer time on writing a relatively precise specification of a system to build attractive, which also helps paving the way to make formal verification for small-scale embeddedsystems more common.
Withthe continuous development of big data and complex softwaresystems, virtual software simulation technology plays an increasingly important role in the fields of data resource search and security prevention appli...
详细信息
the debugging of high-speed trains is a process to ensure the stable operation of trains, and traditional debugging equipment has some problems in terms of comprehensiveness, reliability, and incomplete testing. In or...
详细信息
this study investigated the effectiveness of deep learning models in assessing the reliability of softwaresystems and the application of recurrent neural network algorithms in reliability prediction. A hybrid model c...
详细信息
More and more autonomous robots adapt multi-kernel based software platform. Multiple operating system's kernels run on the same hardware via virtualization to process the various tasks. the microkernel is suitable...
详细信息
暂无评论