Leopoldo Yoshioka

Prof. Leopoldo Yoshioka

IEEE
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
29 October 2023 in Sensors
Verified
Journal Article
Open access
PDF file
RDRodrigo F. Daguano
Rodrigo F. Daguano
LYLeopoldo Yoshioka
Leopoldo Yoshioka
MNMarcio Lobo Netto
Marcio Lobo Netto
CMClaudio L. Marte
Claudio L. Marte
Cassiano Augusto IslerCassiano Augusto Isler
Cassiano Augusto Isler
JJJoao F. Justo
Joao F. Justo
Traffic simulations are valuable tools for urban mobility planning and operation, particularly in large cities. Simulation-based microscopic models have enabled traffic engineers to understand local transit and transport behaviors more deeply and manage urban mobility. However, for the simulations to be effective, the transport network and user behavior parameters must be calibrated to mirror real scenarios. In general, calibration is performed manually by traffic engineers who use their knowledge and experience to adjust the parameters of the simulator. Unfortunately, there is still no systematic and automatic process for calibrating traffic simulation networks, although some methods have been proposed in the literature. This study proposes a methodology that facilitates the calibration process, where an artificial neural network (ANN) is trained to learn the behavior of the transport network of interest. The ANN used is the Multi-Layer Perceptron (MLP), trained with back-propagation methods. Based on this learning procedure, the neural network can select the optimized values of the simulation parameters that best mimic the traffic conditions of interest. Experiments considered two microscopic models of traffic and two psychophysical models (Wiedemann 74 and Wiedemann 99). The microscopic traffic models are located in the metropolitan region of São Paulo, Brazil. Moreover, we tested the different configurations of the MLP (layers and numbers of neurons) as well as several variations of the backpropagation training method: Stochastic Gradient Descent (SGD), Adam, Adagrad, Adadelta, Adamax, and Nadam. The results of the experiments show that the proposed methodology is accurate and efficient, leading to calibration with a correlation coefficient greater than 0.8, when the calibrated parameters generate more visible effects on the road network, such as travel times, vehicle counts, and average speeds. For the psychophysical parameters, in the most simplified model (W74), the correlation coefficient was greater than 0.7. The advantage of using ANN for the automatic calibration of simulation parameters is that it allows traffic engineers to carry out comprehensive studies on a large number of future scenarios, such as at different times of the day, as well as on different days of the week and months of the year.

10 March 2025 in IEEE Access
Verified
Journal Article
Open access
MCMarcus Vinícius Leal De Carvalho
Marcus Vinícius Leal De Carvalho
RSRoberto Simoni
Roberto Simoni
LYLeopoldo Yoshioka
Leopoldo Yoshioka
JJJoão Francisco Justo
João Francisco Justo
BKBruno Mamoru Kato Kawakami
Bruno Mamoru Kato Kawakami
Autonomous vehicles (AVs) are revolutionizing transportation by enhancing safety and efficiency. Despite significant advancements, the literature lacks a systematic benchmarking methodology for open-source autonomous driving stacks (ADS) in realistic simulation environments. This paper presents a methodology for comparing the performance of different autonomous driving (AD) frameworks. It has been applied to compare two leading open-source frameworks, Autoware.AI and Apollo.Auto, focusing on their trajectory planning modules in simulated scenarios. We conducted comprehensive simulations using the LGSVL simulator, integrating both stacks with high-fidelity sensor models and realistic environments. Using a diverse set of performance metrics, including trajectory deviation, computational effort, safety distance, energy consumption, and trajectory smoothness, we could quantitatively assess any planner. Our assessment revealed distinct trade-offs between the planners. Autoware.AI’s Open Planner provided higher safety margins by maintaining larger clearance distances from obstacles, resulting in increased path lengths and energy consumption. Conversely, Apollo.Auto’s Public Road planner offered smoother trajectories and enhanced passenger comfort but required higher computational resources. This work bridges the gap in benchmarking open-source ADS, providing valuable insights into their performance. Our findings contribute to the development of safer and more efficient autonomous vehicles by offering a systematic evaluation of key open-source platforms and highlighting the trade-offs inherent in trajectory planning.

12 November 2024 in Sensors
Verified
Journal Article
Open access
PDF file
DRDaniel Carvalho Ramos
Daniel Carvalho Ramos
LFLucas Reksua Ferreira
Lucas Reksua Ferreira
MSMax Mauro Santos
Max Mauro Santos
ETEvandro Teixeira
Evandro Teixeira
LYLeopoldo Yoshioka
Leopoldo Yoshioka
Asad MalikAsad Malik
Asad Malik
Perception systems for assisted driving and autonomy enable the identification and classification of objects through a concentration of sensors installed in vehicles, including Radio Detection and Ranging (RADAR), camera, Light Detection and Ranging (LIDAR), ultrasound, and HD maps. These sensors ensure a reliable and robust navigation system. Radar, in particular, operates with electromagnetic waves and remains effective under a variety of weather conditions. It uses point cloud technology to map the objects in front of you, making it easy to group these points to associate them with real-world objects. Numerous clustering algorithms have been developed and can be integrated into radar systems to identify, investigate, and track objects. In this study, we evaluate several clustering algorithms to determine their suitability for application in automotive radar systems. Our analysis covered a variety of current methods, the mathematical process of these methods, and presented a comparison table between these algorithms, including Hierarchical Clustering, Affinity Propagation Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Mini-Batch K-Means, K-Means Mean Shift, OPTICS, Spectral Clustering, and Gaussian Mixture. We have found that K-Means, Mean Shift, and DBSCAN are particularly suitable for these applications, based on performance indicators that assess suitability and efficiency. However, DBSCAN shows better performance compared to others. Furthermore, our findings highlight that the choice of radar significantly impacts the effectiveness of these object recognition methods.

26 September 2018 in IEEE Access
Verified
Journal Article
Open access
Bruno Martin de Alcântara DiasBruno Martin de Alcântara Dias
Bruno Martin de Alcântara Dias
ALArmando Antonio Maria Lagana
Armando Antonio Maria Lagana
JJJoao F. Justo
Joao F. Justo
LYLeopoldo Yoshioka
Leopoldo Yoshioka
MSMax Mauro Santos
Max Mauro Santos
Zonghua GuZonghua Gu
Zonghua Gu
A Spark ignition (SI) engine is a complex, multi-domain component of the vehicle powertrain system. The engine control module (ECM) for an SI engine must achieve both high performance and good fuel efficiency. In this paper, we present a model-based development methodology for an open architecture ECM, addressing the entire development lifecycle including a control algorithm design, parameter calibration, hardware/software implementation, and verification/validation of the final system, both with bench tests on a dynamometer and in a real vehicle on the road. The ECM is able to achieve similar performance as the original proprietary ECM provided by the original equipment manufacturer. Its flexible and modular design enables easy extensibility with new control algorithms, and development of new engine types.

16 November 2021 in IEEE Access
Verified
Journal Article
Open access
KHKleber Nogueira Hodel
Kleber Nogueira Hodel
JSJosé Reinaldo Da Silva
José Reinaldo Da Silva
LYLeopoldo Yoshioka
Leopoldo Yoshioka
JJJoao F. Justo
Joao F. Justo
MSMax Mauro Santos
Max Mauro Santos
Road vehicles have incorporated several functionalities over the last decade, with an increasing incorporation of electronic embedded systems. Most of those functionalities are controlled, managed, and supervised by distributed software, within many interconnected Electronic Control Units (ECUs). Within such context, new methodologies for tests of the distributed software functions must be developed to ensure proper performance at the integration level, complying with the requirement specifications. Many strategies have emerged to organize the multiple levels of software testing in automotive embedded systems, to reduce costs and improve its effectiveness and robustness, but there is a need for a methodology to structure, optimize, and plan the tests with real automotive criteria. This work aims to extend the multiple levels of testing concept and propose a method to analyze, design, and evaluate the application system upfront to derive a software testing plan. This method has been developed incorporating the automotive application characteristics, including functional safety requirements specified in the ISO 26262 standard to structure and plan the embedded software test in vehicle development. The proposed method combines an enhancement on plan testing strategy, matching requirement specifications, with adherence to the current methods used in practice by the automotive industry. Such a process will help the automotive industry to follow some concrete steps during validation, to optimize the test volume and facilitate documentation of developed activities, improving the safety and security of automotive systems in the early stages of automotive embedded software, when the details of each function are not yet implemented at the component level.

25 July 2024 in IEEE Transactions on Education
Verified
Journal Article
Closed access
WBWesley Beccaro
Wesley Beccaro
EGElisabete Galeazzo
Elisabete Galeazzo
DCDenise Consonni
Denise Consonni
HPHenrique Peres
Henrique Peres
LYLeopoldo Yoshioka
Leopoldo Yoshioka
Contribution: The evaluation of analog-to-digital conversion methods constitutes a key component of an Instrumentation course. This study introduces an affordable educational platform based on Arduino UNO board designed for teaching analog-to-digital conversion concepts, supported by virtual instruments (VIs). Background: ADCs are electronic devices found in a wide range of consumer electronics, such as smartphones and Internet of Things (IoT) devices. In order to investigate the fundamental aspects of ADCs, a data acquisition system is required. However, high-quality ADC systems tend to be expensive. Alternatively, cost-effective microcontrollers can serve as an educational platform for conducting experimental procedures, including tests, characterization, and calibration. Intended Outcomes: The proposed experiment concentrates on elucidating the theoretical foundations of analog-to-digital conversion, along with providing in-depth insights into the technical details involved in characterizing and calibrating ADCs. Application Design: Four VIs have been developed and are employed to investigate concepts, such as resolution, nonlinearity, aliasing, and to determine offset and gain errors. Findings: The learning experience and the usability of the system were assessed through questionnaires distributed to a total of 105 students. In addition, the final exam was used to assess the performance of 29 students. The results indicate that the students significantly improved their ability to understand, apply, and analyze essential aspects of ADC after engaging in the experiments, demonstrating substantial learning gains.