Radu Grosu

Radu Grosu

Auto-generated profile, based on publicly available information

Radu Grosu

Technische Universität Wien

Research Focus & Metrics

Academic Career Length

28Years (1999-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 October 2020 in Nature Machine Intelligence
Verified
Journal Article
Closed access
PDF file
MLMathias Lechner
Mathias Lechner
RHRamin Hasani
Ramin Hasani
AAAlexander Amini
Alexander Amini
THThomas A. Henzinger
Thomas A. Henzinger
DRDaniela Rus
Daniela Rus
RGRadu Grosu
Radu Grosu

1 March 2022 in Renewable and Sustainable Energy Reviews
Verified
Journal Article
Closed access
PLPenghua Li
Penghua Li
ZZZijian Zhang
Zijian Zhang
RGRadu Grosu
Radu Grosu
Zhongwei DengZhongwei Deng
Zhongwei Deng
JHJie Hou
Jie Hou
RWRui Wu
Rui Wu

10 September 2024 in IEEE Transactions on Transportation Electrification
Verified
Journal Article
Closed access
PLPenghua Li
Penghua Li
XWXiankui Wu
Xiankui Wu
RGRadu Grosu
Radu Grosu
JHJie Hou
Jie Hou
MIMamadsho Ilolov
Mamadsho Ilolov
SXSheng Xiang
Sheng Xiang
In recent years, artificial neural networks (ANNs) have significantly advanced in both health estimation and lifetime prediction of lithium-ion batteries. The great success of ANNs stems primarily from their scalability in encoding large-scale data and manoeuvre billions of model parameters. However, there are still many challenges in balancing predictive accuracy and deployment feasibility. For instance, shallow ANNs are often more efficient but may sometimes sacrifice accuracy, whereas deep hybrid ANNs often achieve strong generalization capabilities, this comes with the trade-off of increased computational demands. To this end, this paper presents a comprehensive survey of ANN-based paradigms for estimating State-of-Health (SOH) and predicting the Remaining-Useful-Life (RUL) of lithium-ion batteries. It covers battery aging mechanisms, available datasets, network architecture, training schemes, advanced machine learning (AML) algorithms, and performance comparison. Furthermore, challenges in battery health diagnosis are reviewed in detail, and comments on future research prospects are discussed and forwarded.

16 June 2021 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
PLPenghua Li
Penghua Li
YYYifeng Yang
Yifeng Yang
RGRadu Grosu
Radu Grosu
GWGuodong WANG
Guodong WANG
RLRui Li
Rui Li
ZHZeng Huang
Zeng Huang
This study proposes a lightweight convolutional neural network with an octave-like convolution mixed block, called OLCMNet, for detecting driver distraction under a limited computational budget. The OLCM block uses point-wise convolution (PC) to expand feature maps into two sets of branches. In the low-frequency branches, we perform average pooling, depth-wise convolution (DC), and upsampling to obtain a low-resolution low-frequency feature map, reducing spatial redundancy and connection density. In the high-frequency branches, the expanded feature map with the original resolution is fed to the DC operator, gaining an apposite receptive field to capture fine details. The feature concatenation of the low-frequency and high-frequency branches is encoded sequentially by a squeeze-and-excitation (SE) module and PC operator, realizing feature global information fusion. Introducing another SE module at the last stage, the OLCMNet facilitates further sensitive information exchange between layers. In addition, with an augmented reality head-up display (ARHUD) platform, we create a Lilong Distracted Driving Behavior (LDDB) Dataset through a series of on-road experiments. Such a dataset contains 14808 videos collected from an infrared camera, covering six driving behaviors of 2468 participants. We manually annotate these videos at five frames per second, obtaining a total of 267378 images. Compared with the existing methods, the embedded hardware platform experiments indicate that OLCMNet hits acceptable trade-offs, namely, 89.53% accuracy for StateFarm Dataset and 95.98% accuracy LDDB Dataset when the latency is 32.8 ± 4.6ms.

27 March 2018 in Formal Methods in System Design
Verified
Journal Article
Open access
PDF file
SJStefan Jakšić
Stefan Jakšić
Ezio BartocciEzio Bartocci
Ezio Bartocci
RGRadu Grosu
Radu Grosu
TNThang Nguyen
Thang Nguyen
DNDejan Nickovic
Dejan Nickovic
In cyber-physical systems (CPS), physical behaviors are typically controlled by digital hardware. As a consequence, continuous behaviors are discretized by sampling and quantization prior to their processing. Quantifying the similarity between CPS behaviors and their specification is an important ingredient in evaluating correctness and quality of such systems. We propose a novel procedure for measuring robustness between digitized CPS signals and signal temporal logic (STL) specifications. We first equip STL with quantitative semantics based on the weighted edit distance, a metric that quantifies both space and time mismatches between digitized CPS behaviors. We then develop a dynamic programming algorithm for computing the robustness degree between digitized signals and STL specifications. In order to promote hardware-based monitors we implemented our approach in FPGA. We evaluated it on automotive benchmarks defined by research community, and also on realistic data obtained from magnetic sensor used in modern cars.

16 October 2018 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
SJStefan Jakšić
Stefan Jakšić
Ezio BartocciEzio Bartocci
Ezio Bartocci
RGRadu Grosu
Radu Grosu
DNDejan Ničković
Dejan Ničković
Runtime verification (RV) is a pragmatic and scalable, yet rigorous technique, to assess the correctness of complex systems, including cyber-physical systems (CPSs). Modern RV tools also allow to measure the distance of a CPS behavior from a given formal requirement, thus, to quantify the robustness of a CPS with respect to perturbations caused by the physical environment. In this paper, we propose algebraic RV (ARV), a general, semantic framework for correctness and robustness monitoring. ARV implements an abstract monitoring procedure, in which the specification language (STL) can be instantiated with various qualitative and quantitative semantics. This allows us to expose the core aspects of RV, by separating the monitoring algorithm from the concrete choice of the STL and its semantics. We demonstrate the effectiveness of our framework on two examples from the automotive domain.