| MA SHANPENG | M, 2回目発表 | インタラクティブメディア設計学 | 加藤 博一, | 清川 清, | 神原 誠之, | 藤本 雄一郎, | 澤邊 太志 | |
|
title: Improving Passenger Trust And Acceptance in Autonomous Vehicles through an Explainer Model
abstract: This research focuses on improving passenger comfort and trust in autonomous vehicles by improving the transparency of the system. Autonomous Driving Systems (ADS) are often perceived as a "black box", making it challenging for passengers to anticipate the vehicle's actions. This unpredictability can cause discomfort and hesitancy in using autonomous vehicles. Although existing research on autonomous vehicles primarily emphasizes safety and efficiency, there is a notable gap in studies addressing passenger trust and acceptance. To address this gap, the study introduces an "Explainer" model that utilizes Augmented Reality (AR) to present structured data from the ADS in a format that is easily comprehensible to users. The model organizes data from the ADS Perception, Planning, and Control modules into causal and intentional information. Causal information illustrates the physical cause-and-effect relationship between the vehicle's movements and its surroundings, while intentional information conveys the vehicle's objectives and intended actions. By offering passengers a clear understanding of the vehicle's behavior, the model aims to bolster their feelings of safety and control, leading to an enhanced sense of comfort and trust in autonomous vehicles. language of the presentation: English | ||||||||
| RAHMAN MD MUSTAFIZUR | M, 1回目発表 | インタラクティブメディア設計学 | 加藤 博一, | 清川 清, | 神原 誠之, | 藤本 雄一郎, | 澤邊 太志 | 山本 豪志朗 |
|
Title: Experience Augmentation in Physical Therapy by Simulating Patient-Specific Walking Motions Abstract: Traditionally, Physical Therapists (PTs) are trained using textbooks, videos, and hands-on experience with real patients to develop their diagnostic and therapeutic skills. However, there is a growing interest in incorporating advanced technologies to enhance this training. This research proposes the integration of generative AI and augmented reality (AR) to create interactive, patient-specific 3D simulations of walking motions, providing a novel tool for gait training. The goal is to develop an AR-based simulator that delivers real-time motion analysis and feedback, allowing PTs to interact with dynamic 3D walking models in an immersive environment. This approach aims to improve diagnostic accuracy and therapeutic skills through experiential learning. The study investigates two primary questions: how can generative AI and AR be effectively combined to generate accurate, patient-specific 3D walking simulations, and what advantages does immersive AR-based gait training offer over traditional training methods? By addressing these questions, the research seeks to demonstrate that AR-enhanced gait training can significantly improve diagnostic precision and therapeutic outcomes, offering a more efficient and patient-centered approach to rehabilitation. Language of the presentation: English | ||||||||
| TUWAEMUESA THAPAKORN | M, 1回目発表 | インタラクティブメディア設計学 | 加藤 博一, | 清川 清, | 神原 誠之, | 藤本 雄一郎, | 澤邊 太志 | |
|
title: Content-driven gaze behavior training system in communication tasks with a virtual character
abstract: Eye gaze behavior is crucial for social performance and communication during interpersonal interactions. Virtual reality (VR) technology is being utilized to develop training systems for enhancing this behavior. However, current training content primarily focuses on increasing the amount of mutual gaze. This study leverages advancements in virtual reality (VR) technology and the use of virtual characters to enhance gaze behaviors in terms of duration, focus areas, and appropriate timing. By immersing participants in controlled VR environments with generative conversational partners, we aim to refine eye gaze behaviors based on general patterns that facilitate effective communication and positive social outcomes. language of the presentation: English | ||||||||
| DANG PHUONG NAM | M, 1回目発表 | ヒューマンロボティクス | 和田 隆広, | 清川 清, | 織田 泰彰, | 劉 海龍 | ||
|
title: Development of an external HMI for autonomous personal mobility vehicles considering passengers' personalities.
abstract: An autonomous personal mobility vehicle (APMV) is a miniaturized autonomous vehicle designed for shared space mobility for everyone. Besides the interaction between APMVs and pedestrians, the interaction between APMVs and passengers is also a critical aspect to consider when designing an external human-machine interface (eHMI). This study introduces an eHMI design method based on human personalities, where the output text and voice for the eHMI that is tailored to the personality of the passenger by utilizing the big five personality traits as input to a deep generative model together with interaction scenarios between APMV and pedestrians. This presentation highlights the relationship between people's personalities and preferences for talking content with the same voice condition. language of the presentation: English | ||||||||
| BORHAN ROWNAK | M, 1回目発表 | サイバーレジリエンス構成学 | 門林 雄基, | 和田 隆広, | 妙中 雄三, | HOSSAIN, Md Delwar | ||
|
title:Cryptographic Approaches for Securing IoT and Robotics in the
Post-Quantum Age
abstract:Quantum computing is rapidly approaching practical implementation, offering unprecedented processing power and problem-solving capabilities across diverse domains. However, this advancement also poses a significant challenge to the security of cryptographic systems. In response to this fact, there has been a push for standardizing cryptographic protocols that can withstand quantum attacks. Through a rigorous selection process, the National Institute of Standards and Technology (NIST) has officially endorsed CRYSTALS-Dilithium as a leading lattice-based digital signature algorithm. While progress has been made in addressing computational challenges in polynomial multiplication, there persists a gap in research concerning hashing enhancements within CRYSTALS-Dilithium key generation. Therefore, developing a robust and secure key generation process for CRYSTALS-Dilithium, especially for Internet of Things (IoT) applications, is crucial. To overcome the limitation of key generation, we introduced an optimized hash function called IOSHA, which improves security compared to SHAKE-256 by utilizing dynamic parameters and constants. Integrating IOSHA into the key generation algorithm of CRYSTALS-Dilithium offers substantial improvements over traditional methods. This integration accelerates processing times while enhancing security against potential threats and maintaining resistance to MLWE-based hard problems without requiring hardware upgrades. language of the presentation: English. | ||||||||