Dai Watanabe

Prof. Dai Watanabe

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
19 January 2024 in Journal of the Brazilian Society of Mechanical Sciences and Engineering
Verified
Journal Article
Closed access
PDF file
PBPhittayut Bunsri
Phittayut Bunsri
SLSorrawit Lophisarn
Sorrawit Lophisarn
PJPattaramon Jongpradist
Pattaramon Jongpradist
Suphanut KongwatSuphanut Kongwat
Suphanut Kongwat
DWDai Watanabe
Dai Watanabe

15 May 2025 in IEEE Access
Verified
Journal Article
Open access
TLThai la
Thai la
LTLinh TAO
Linh TAO
DWDai Watanabe
Dai Watanabe
Despite the significant advancements in learning-based stereo matching algorithms, a significant challenge remains: the high computational cost and memory demands of 3D convolutions, which hinder real-time deployment on resource-constrained platforms like edge devices. In this paper, we present a novel approach that completely avoids the use of 3D convolutions, with the goal of achieving faster inference speeds while maintaining comparable accuracy to existing state-of-the-art methods. Our proposed solution revolves around a 2D cost aggregation technique, which serves as a viable alternative to traditional 3D convolutions, delivering similar results in terms of accuracy. This innovative method significantly reduces computational overhead, facilitating more efficient resource utilization and paving the way for real-time applications. Complementing our 2D cost aggregation module, we introduce a multi-stage feature extractor, designed to enhance feature representation while remaining straightforward and lightweight. The integration of the 2D cost aggregation and multi-stage feature extraction results in an efficient architecture for cost aggregation, simplifying the model and ensuring computational efficiency without sacrificing accuracy. This framework delivers high-performance stereo matching suitable for devices with limited computational capabilities. Through evaluation on benchmark datasets, we demonstrate the effectiveness of our approach, highlighting its potential for real-time 3D perception in applications. By addressing the constraints imposed by 3D convolutions and offering a more pragmatic alternative, this work bridges the gap between high-performance stereo matching algorithms and the realities of edge computing environments.

30 January 2026 in IEEE Access
Verified
Journal Article
Open access
TLThai la
Thai la
LTLinh TAO
Linh TAO
DADao Tuan Anh
Dao Tuan Anh
DWDai Watanabe
Dai Watanabe
Adverse weather (rain, fog, snow) corrupts LiDAR through backscatter and attenuation, producing near-range clutter and far-range sparsity that degrade downstream perception and planning. Classical geometric filters depend on hand-tuned thresholds and often erode valid structure at long range; recent learning methods improve accuracy but are heavy for embedded deployment. We present LRINet, a straightforward and efficient range-image network for denoising LiDAR point clouds in adverse weather. LRINet operates entirely in a 2-D cylindrical projection. This approach preserves sensor topology and ensures computational efficiency. The network pairs depth-guided attention with a coordinate encoder and a global-context block. It then uses an attention-gated decoder for the final output. Despite its minimalist design, LRINet delivers competitive performance against state-of-the-art methods across rainy, foggy, and snowy conditions while maintaining a markedly smaller memory and latency footprint suitable for real-time on-board use. Experiments on climate-chamber rain/fog data and semi-synthetic snowy driving data show a practical balance of accuracy, robustness, and throughput, making LRINet a drop-in preprocessing module for safety-critical autonomy stacks.

1 January 2026 in Journal of Climbing and Walking Robots
Verified
Journal Article
Open access
PDF file
TWTomohiro Watanabe
Tomohiro Watanabe
DWDai Watanabe
Dai Watanabe
KIKojiro Iizuka
Kojiro Iizuka
Numerous missions have targeted planetary bodies, highlighting the need for advanced mobility technologies that enable exploration beyond the surface. Vibration-driven underground locomotion is a promising approach because vibration can alter soil behavior and reduce resistance to movement. However, the particle-scale mechanisms responsible for this reduction remain unclear, limiting the development of efficient vibration-based locomotion systems. In this study, particle-based numerical simulations and experiments were used to investigate how vibration reduces the resistance acting on an object moving through granular ground. A numerical model was developed to reproduce rod-dragging experiments conducted with a vibration amplitude of 0.25 mm at frequencies of 30 Hz and 50 Hz. The simulated resistance showed good agreement with the experimental results under the no-vibration and 50 Hz conditions, with errors within 2%. The results demonstrated that vibration reduced the resistance acting on the moving rod compared with the no-vibration condition. Particle-scale analysis further revealed that vibration generated high-velocity particle regions ahead of the rod and promoted particle rearrangement around it. These effects suppressed particle accumulation in front of the rod, thereby reducing resistance to motion. The findings clarify the relationship between vibration-induced particle flow and reduced soil resistance, providing a physical basis for the development of vibration-driven underground locomotion systems. Beyond planetary exploration, the numerical modeling of vibration-induced changes in granular ground may also contribute to broader engineering applications, including civil engineering problems involving vibrated granular materials. Received: 27 March 2026 | Revised: 27 June 2026 | Accepted: 2 September 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data are available from the corresponding author upon reasonable request. Author Contribution Statement Tomohiro Watanabe: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Visualization, Project administration, Funding acquisition. Dai Watanabe: Conceptualization, Resources, Writing – review & editing. Kojiro Iizuka: Conceptualization, Resources, Writing – review & editing, Supervision.

11 October 2024 in Mathematical Modelling and Analysis
Verified
Journal Article
Open access
PDF file
DWDai Watanabe
Dai Watanabe
SYShuji Yoshikawa
Shuji Yoshikawa
To improve ride comfort, an oil damper should restrain its damping force in a high-velocity range. A lot of dampers with such properties were developed. For example, the one controlling the flux of oil by a leaf valve is widely adapted and reasonable. However, it is difficult to represent its dynamics with a simple mathematical model, and the cost of a computational fluid dynamics is too expensive. To overcome the disadvantages, the first author in [15] developed the other mechanical oil damper with sub-pistons instead of the leaf valve, which enabled us to propose a simple mathematical model with linear ordinary differential equations, thanks to the simple mechanism to control the oil flow. In this article, we give a more detailed mathematical model for the damper taking the dynamic pressure resistance into account, which is represented by nonlinear ordinary differential equations. In addition, a numerical scheme for the model is also proposed and its mathematical analysis such as the validity of the numerical solutions is shown.

11 March 2026
Verified
Conference Paper
Open access
TLThai La
Thai La
LTLinh TAO
Linh TAO
NHNguyen Canh Van Ha
Nguyen Canh Van Ha
DWDai Watanabe
Dai Watanabe