Hiroshi Yasuda
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hiroshi Yasuda has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Fanjun Bu
Computer Science, Cornell University, Cornell Tech, New York, New York, USA
2 shared publications Congkai Shen
1 shared publicationDavid Goedicke
1 shared publication- Gyanendra Sharma
Toyota Research Institute 94022 Los Altos CA
1 shared publication - Gyanendra Sharma1 shared publication
- Gyanendra Sharma
Toyota Research Institute Los Altos 94022 CA
1 shared publication
Latest Publications
- Open Access
Spatial Envelope MPC: High Performance Driving without a Reference
This paper presents a novel envelope based model predictive control (MPC) framework designed to enable autonomous vehicles to handle high performance driving across a wide range of scenarios without a predefined reference. In high performance autonomous driving, safe operation at the vehicle's dynamic limits requires a real time planning and control framework capable of accounting for key vehicle dynamics and environmental constraints when following a predefined reference trajectory is suboptimal or even infeasible. State of the art planning and control frameworks, however, are predominantly reference based, which limits their performance in such situations. To address this gap, this work first introduces a computationally efficient vehicle dynamics model tailored for optimization based control and a continuously differentiable mathematical formulation that accurately captures the entire drivable envelope. This novel model and formulation allow for the direct integration of dynamic feasibility and safety constraints into a unified planning and control framework, thereby removing the necessity for predefined references. The challenge of envelope planning, which refers to maximally approximating the safe drivable area, is tackled by combining reinforcement learning with optimization techniques. The framework is validated through both simulations and real world experiments, demonstrating its high performance across a variety of tasks, including racing, emergency collision avoidance and off road navigation. These results highlight the framework's scalability and broad applicability across a diverse set of scenarios.
Open AccessSpatial Envelope MPC: High Performance Driving without a Reference
This paper presents a novel envelope based model predictive control (MPC) framework designed to enable autonomous vehicles to handle high performance driving across a wide range of scenarios without a predefined reference. In high performance autonomous driving, safe operation at the vehicle's dynamic limits requires a real time planning and control framework capable of accounting for key vehicle dynamics and environmental constraints when following a predefined reference trajectory is suboptimal or even infeasible. State of the art planning and control frameworks, however, are predominantly reference based, which limits their performance in such situations. To address this gap, this work first introduces a computationally efficient vehicle dynamics model tailored for optimization based control and a continuously differentiable mathematical formulation that accurately captures the entire drivable envelope. This novel model and formulation allow for the direct integration of dynamic feasibility and safety constraints into a unified planning and control framework, thereby removing the necessity for predefined references. The challenge of envelope planning, which refers to maximally approximating the safe drivable area, is tackled by combining reinforcement learning with optimization techniques. The framework is validated through both simulations and real world experiments, demonstrating its high performance across a variety of tasks, including racing, emergency collision avoidance and off road navigation. These results highlight the framework's scalability and broad applicability across a diverse set of scenarios.
- Open Access
Boosting Visual Fidelity in Driving Simulations through Diffusion Models
Diffusion models have made substantial progress in facilitating image generation and editing. As the technology matures, we see its potential in the context of driving simulations to enhance the simulated experience. In this paper, we explore this potential through the introduction of a novel system designed to boost visual fidelity. Our system, DRIVE (Diffusion-based Realism Improvement for Virtual Environments), leverages a diffusion model pipeline to give a simulated environment a photorealistic view, with the flexibility to be adapted for other applications. We conducted a preliminary user study to assess the system's effectiveness in rendering realistic visuals and supporting participants in performing driving tasks. Our work not only lays the groundwork for future research on the integration of diffusion models in driving simulations but also provides practical guidelines and best practices for their application in this context.
Open AccessBoosting Visual Fidelity in Driving Simulations through Diffusion Models
Diffusion models have made substantial progress in facilitating image generation and editing. As the technology matures, we see its potential in the context of driving simulations to enhance the simulated experience. In this paper, we explore this potential through the introduction of a novel system designed to boost visual fidelity. Our system, DRIVE (Diffusion-based Realism Improvement for Virtual Environments), leverages a diffusion model pipeline to give a simulated environment a photorealistic view, with the flexibility to be adapted for other applications. We conducted a preliminary user study to assess the system's effectiveness in rendering realistic visuals and supporting participants in performing driving tasks. Our work not only lays the groundwork for future research on the integration of diffusion models in driving simulations but also provides practical guidelines and best practices for their application in this context.
- Open Access
Portobello: Extending Driving Simulation from the Lab to the Road
In automotive user interface design, testing often starts with lab-based driving simulators and migrates toward on-road studies to mitigate risks. Mixed reality (XR) helps translate virtual study designs to the real road to increase ecological validity. However, researchers rarely run the same study in both in-lab and on-road simulators due to the challenges of replicating studies in both physical and virtual worlds. To provide a common infrastructure to port in-lab study designs on-road, we built a platform-portable infrastructure, Portobello, to enable us to run twinned physical-virtual studies. As a proof-of-concept, we extended the on-road simulator XR-OOM with Portobello. We ran a within-subjects, autonomous-vehicle crosswalk cooperation study (N=32) both in-lab and on-road to investigate study design portability and platform-driven influences on study outcomes. To our knowledge, this is the first system that enables the twinning of studies originally designed for in-lab simulators to be carried out in an on-road platform.
Open AccessPortobello: Extending Driving Simulation from the Lab to the Road
In automotive user interface design, testing often starts with lab-based driving simulators and migrates toward on-road studies to mitigate risks. Mixed reality (XR) helps translate virtual study designs to the real road to increase ecological validity. However, researchers rarely run the same study in both in-lab and on-road simulators due to the challenges of replicating studies in both physical and virtual worlds. To provide a common infrastructure to port in-lab study designs on-road, we built a platform-portable infrastructure, Portobello, to enable us to run twinned physical-virtual studies. As a proof-of-concept, we extended the on-road simulator XR-OOM with Portobello. We ran a within-subjects, autonomous-vehicle crosswalk cooperation study (N=32) both in-lab and on-road to investigate study design portability and platform-driven influences on study outcomes. To our knowledge, this is the first system that enables the twinning of studies originally designed for in-lab simulators to be carried out in an on-road platform.
- Open Access
Detection Threshold of Audio Haptic Asynchrony in a Driving Context
In order to provide perceptually accurate multimodal feedback during driving situations, it is vital to understand the threshold at which drivers are able to recognize asyncrony between multiple incoming Stimuli. In this work, we investigated and report the \textit{detection threshold} (DT) of asynchrony between audio and haptic feedback, in the context of a force feedback steering wheel. We designed the experiment to loosely resemble a driving situation where the haptic feedback was provided through a steering wheel (\textit{Sensodrive}), while the accompanying audio was played through noise cancelling headphones. Both feedbacks were designed to resemble rumble strips, that are generally installed on the side of major roadways as a safety tool. The results indicate that, for $50\%$ of the participants, asynchrony was detectable outside the range of -75 ms and 110 ms, where the former is related to perceiving audio before haptic and vice versa for the latter. We were also able to concur with previous studies, which state that latency is perceivable at a lower threshold when audio precedes haptic stimuli.Open AccessDetection Threshold of Audio Haptic Asynchrony in a Driving Context
In order to provide perceptually accurate multimodal feedback during driving situations, it is vital to understand the threshold at which drivers are able to recognize asyncrony between multiple incoming Stimuli. In this work, we investigated and report the \textit{detection threshold} (DT) of asynchrony between audio and haptic feedback, in the context of a force feedback steering wheel. We designed the experiment to loosely resemble a driving situation where the haptic feedback was provided through a steering wheel (\textit{Sensodrive}), while the accompanying audio was played through noise cancelling headphones. Both feedbacks were designed to resemble rumble strips, that are generally installed on the side of major roadways as a safety tool. The results indicate that, for $50\%$ of the participants, asynchrony was detectable outside the range of -75 ms and 110 ms, where the former is related to perceiving audio before haptic and vice versa for the latter. We were also able to concur with previous studies, which state that latency is perceivable at a lower threshold when audio precedes haptic stimuli. - Open Access
Continuous Visual Feedback of Risk for Haptic Lateral Assistance
Lateral assistance systems are a significant part of the currently existing Advanced Driving Assistance System (ADAS). In most cases, such systems provide intermittent audio and haptic feedback rather than continuous feedback, making it difficult for drivers to form a clear mental model. In this paper, we propose continuous visual feedback for the lateral assistance system by leveraging the Head-Up Display (HUD) alongside haptic feedback through the steering wheel. The HUD provides visualization of the risk profile underlying the haptic feedback. We hypothesize that our proposed visualization helps form a clear mental model and improves the system's acceptance. We conduct a user study on a simulated version of a car driving on a two-lane freeway and compare the haptic lateral assistance system with and without the visualization on the HUD. While both conditions received high acceptance scores, there was no significant gain or deterioration in acceptance between them. We discuss potential improvements in the visualization based on anticipation performance and qualitative feedback from users.
Open AccessContinuous Visual Feedback of Risk for Haptic Lateral Assistance
Lateral assistance systems are a significant part of the currently existing Advanced Driving Assistance System (ADAS). In most cases, such systems provide intermittent audio and haptic feedback rather than continuous feedback, making it difficult for drivers to form a clear mental model. In this paper, we propose continuous visual feedback for the lateral assistance system by leveraging the Head-Up Display (HUD) alongside haptic feedback through the steering wheel. The HUD provides visualization of the risk profile underlying the haptic feedback. We hypothesize that our proposed visualization helps form a clear mental model and improves the system's acceptance. We conduct a user study on a simulated version of a car driving on a two-lane freeway and compare the haptic lateral assistance system with and without the visualization on the HUD. While both conditions received high acceptance scores, there was no significant gain or deterioration in acceptance between them. We discuss potential improvements in the visualization based on anticipation performance and qualitative feedback from users.