Takato Horii
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Takato Horii has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Takayuki Nagai
Department of Systems Innovation, Graduate School of Engineering Science, Osaka University, Osaka, Japan
6 shared publications - Hiroki Ishizuka
Osaka University
3 shared publications - Sei Ikeda
Graduate School of Engineering Science, Osaka University, Osaka, Japan
3 shared publications - Takumi Kawasetsu
Graduate School of Engineering, Kyoto University, Kyoto, Japan
3 shared publications - Xiaoxu Feng
Graduate School of Engineering Science, The University of Osaka, Osaka, Japan
3 shared publications - Kazuki Miyazawa
Graduate School of Engineering Science, Osaka University, Osaka, Japan
2 shared publications
Latest Publications
- Open Access
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (11) , 13549-13556
SADP: Subgoal-Aware Diffusion Policy for Long-Horizon Manipulation Learned From Foundation Model Generated Demonstrations
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (11) , 13549-13556
Open AccessSADP: Subgoal-Aware Diffusion Policy for Long-Horizon Manipulation Learned From Foundation Model Generated Demonstrations
- Open Access
- 21 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 78614-78629
Distance-Aware World Model-Based Reinforcement Learning for Mobile Manipulation Behaviors
Separate control of mobile manipulation offers a promising paradigm for managing its complexity by planning for these modalities independently. However, end-to-end learning of such behaviors necessitates effective mechanisms for both embodiment selection and motion planning. For low-level motion planning, prior world model-based reinforcement learning (RL) approaches have employed the similarity of latent states as a dense reward proxy. However, using latent similarity as an implicit distance metric yields inaccurate representations and depends heavily on predefined goal states. In this work, we address the limitations of similarity-based rewards by explicitly integrating 3D spatial information into the world model and deriving more accurate rewards from that distance-aware model for policy training. For high-level embodiment selection, prior model-based methods have either utilized reachability rewards inferred from predicted task rewards or relied on heuristic composite rewards. In this study, we retain the reachability reward and demonstrate that it yields superior performance in embodiment selection compared with composite reward formulations. By combining reachability-based selection and distance-guided planning, the model improves the performance of end-to-end learning for separate mobile manipulation behaviors. Extensive evaluations were conducted across two simulation environments with both fixed and randomized target settings. For motion planning, the distance reward achieves near 100% success rates across all configurations, significantly outperforming the baseline similarity reward, which achieves only 60% success in fixed-target scenarios and drops below 20% when targets are randomized. Regarding embodiment selection, the reachability-based separate control rapidly achieves a stable result with over 85% of arm activations occurring rationally within the near-goal region. Conversely, the heuristic composite reward slowly increases to a maximum of 60% for near-goal arm selections. These findings validate that the reachability reward provides a more effective and efficient mechanism for embodiment selection.
- 21 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 78614-78629
Open AccessDistance-Aware World Model-Based Reinforcement Learning for Mobile Manipulation Behaviors
Separate control of mobile manipulation offers a promising paradigm for managing its complexity by planning for these modalities independently. However, end-to-end learning of such behaviors necessitates effective mechanisms for both embodiment selection and motion planning. For low-level motion planning, prior world model-based reinforcement learning (RL) approaches have employed the similarity of latent states as a dense reward proxy. However, using latent similarity as an implicit distance metric yields inaccurate representations and depends heavily on predefined goal states. In this work, we address the limitations of similarity-based rewards by explicitly integrating 3D spatial information into the world model and deriving more accurate rewards from that distance-aware model for policy training. For high-level embodiment selection, prior model-based methods have either utilized reachability rewards inferred from predicted task rewards or relied on heuristic composite rewards. In this study, we retain the reachability reward and demonstrate that it yields superior performance in embodiment selection compared with composite reward formulations. By combining reachability-based selection and distance-guided planning, the model improves the performance of end-to-end learning for separate mobile manipulation behaviors. Extensive evaluations were conducted across two simulation environments with both fixed and randomized target settings. For motion planning, the distance reward achieves near 100% success rates across all configurations, significantly outperforming the baseline similarity reward, which achieves only 60% success in fixed-target scenarios and drops below 20% when targets are randomized. Regarding embodiment selection, the reachability-based separate control rapidly achieves a stable result with over 85% of arm activations occurring rationally within the near-goal region. Conversely, the heuristic composite reward slowly increases to a maximum of 60% for near-goal arm selections. These findings validate that the reachability reward provides a more effective and efficient mechanism for embodiment selection.
- 28 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (11) , 17634-17646
External Photoreflective Tactile Sensing Based on Surface Deformation Measurement
We present a tactile-sensing method enabled by the mechanical compliance of soft robots: an externally attachable photoreflective module reads surface deformation of silicone skin to estimate contact force while (i) locating the sensing module off the contact interface, (ii) keeping the skin free of embedded transducers/wiring, and (iii) enabling rapid replacement and reconfiguration of the module. Locating the sensor off the contact interface reduces damage risk, and the non-embedded architecture preserves softness while simplifying fabrication and maintenance through a plug-in, replaceable module. We first characterize the optical sensing element and the compliant skin, then determine the design of a prototype tactile sensor. Compression experiments validate the approach, exhibiting a monotonic force–output relationship consistent with theory, low hysteresis, high repeatability over repeated cycles, and small response lag over 0.1–10 mm/s indentation speeds. We further demonstrate integration on a soft robotic gripper, where the module reliably detects grasp events. Compared with liquid-filled or wire-embedded tactile skins, the proposed modular “add-on” architecture enhances durability, reduces wiring complexity, and supports easy redeployment across diverse robot geometries without disassembling the robot skin. Because the sensing principle reads skin-strain patterns, it also suggests extensions to other somatosensory cues—such as join-tangle or actuator-state estimation—from surface deformation. Overall, leveraging surface compliance with an external optical module provides a practical and robust route to equip soft robots with force perception while preserving structural flexibility and manufacturability, paving the way for robotic applications and safe human–robot collaboration.
- 28 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (11) , 17634-17646
External Photoreflective Tactile Sensing Based on Surface Deformation Measurement
We present a tactile-sensing method enabled by the mechanical compliance of soft robots: an externally attachable photoreflective module reads surface deformation of silicone skin to estimate contact force while (i) locating the sensing module off the contact interface, (ii) keeping the skin free of embedded transducers/wiring, and (iii) enabling rapid replacement and reconfiguration of the module. Locating the sensor off the contact interface reduces damage risk, and the non-embedded architecture preserves softness while simplifying fabrication and maintenance through a plug-in, replaceable module. We first characterize the optical sensing element and the compliant skin, then determine the design of a prototype tactile sensor. Compression experiments validate the approach, exhibiting a monotonic force–output relationship consistent with theory, low hysteresis, high repeatability over repeated cycles, and small response lag over 0.1–10 mm/s indentation speeds. We further demonstrate integration on a soft robotic gripper, where the module reliably detects grasp events. Compared with liquid-filled or wire-embedded tactile skins, the proposed modular “add-on” architecture enhances durability, reduces wiring complexity, and supports easy redeployment across diverse robot geometries without disassembling the robot skin. Because the sensing principle reads skin-strain patterns, it also suggests extensions to other somatosensory cues—such as join-tangle or actuator-state estimation—from surface deformation. Overall, leveraging surface compliance with an external optical module provides a practical and robust route to equip soft robots with force perception while preserving structural flexibility and manufacturability, paving the way for robotic applications and safe human–robot collaboration.
- Open Access
- 17 March 2026
- Published by Springer Nature
- p. 97-103
Emotion and Predictive Processing: How Are Emotions Made?
- 17 March 2026
- Published by Springer Nature
- p. 97-103
Open AccessEmotion and Predictive Processing: How Are Emotions Made?
- Open Access
- 17 March 2026
- Published by Springer Nature
- p. 91-96
Curiosity and Exploration: Why Do We Want to Learn?
- 17 March 2026
- Published by Springer Nature
- p. 91-96
Open AccessCuriosity and Exploration: Why Do We Want to Learn?
- Open Access
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 20905-20915
LocoGPT: GPT-Based Multi-Humanoid-Task Policy for Humanoid Locomotion
Recent studies have explored developing controllers that can generalize across several humanoid robots that differ in shape and size. However, limited studies have developed cross-embodiment policies for humanoids in a low data regime with varied observation and action spaces. In this study, we propose a GPT-based multi-humanoid locomotion policy called LocoGPT, (where GPT refers to Generative Pre-Trained Transformer), trained on multiple humanoids with varied observation and action spaces that can generate various locomotion behaviors for durations that touch close to one minute on an average. From our evaluations, we show that LocoGPT has high cross-embodiment learning capability and can generalize locomotion behaviors like walking, running, and carrying across different humanoids better than the baselines. LocoGPT outperforms the Multi-Layer Perceptron (MLP) baselines in generating actions for locomotion-based tasks, and can do so by leveraging body specific modules with a shared actor network in a low data regime. We further show that the motions generated by LocoGPT are stable and relatively close to expert data using metrics such as action rate and joint acceleration and jerk smoothness ratios. Additionally, we condense the model as much as possible and show that we can retain similar performance levels.
- 6 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 20905-20915
Open AccessLocoGPT: GPT-Based Multi-Humanoid-Task Policy for Humanoid Locomotion
Recent studies have explored developing controllers that can generalize across several humanoid robots that differ in shape and size. However, limited studies have developed cross-embodiment policies for humanoids in a low data regime with varied observation and action spaces. In this study, we propose a GPT-based multi-humanoid locomotion policy called LocoGPT, (where GPT refers to Generative Pre-Trained Transformer), trained on multiple humanoids with varied observation and action spaces that can generate various locomotion behaviors for durations that touch close to one minute on an average. From our evaluations, we show that LocoGPT has high cross-embodiment learning capability and can generalize locomotion behaviors like walking, running, and carrying across different humanoids better than the baselines. LocoGPT outperforms the Multi-Layer Perceptron (MLP) baselines in generating actions for locomotion-based tasks, and can do so by leveraging body specific modules with a shared actor network in a low data regime. We further show that the motions generated by LocoGPT are stable and relatively close to expert data using metrics such as action rate and joint acceleration and jerk smoothness ratios. Additionally, we condense the model as much as possible and show that we can retain similar performance levels.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 173163-173179
Offline World Model-Based Reinforcement Learning for Mobile Manipulation Behaviors
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 173163-173179
Open AccessOffline World Model-Based Reinforcement Learning for Mobile Manipulation Behaviors
- Open Access
External Photoreflective Tactile Sensing Based on Surface Deformation Measurement
We present a tactile sensing method enabled by the mechanical compliance of soft robots; an externally attachable photoreflective module reads surface deformation of silicone skin to estimate contact force without embedding tactile transducers. Locating the sensor off the contact interface reduces damage risk, preserves softness, and simplifies fabrication and maintenance. We first characterize the optical sensing element and the compliant skin, thendetermine the design of a prototype tactile sensor. Compression experiments validate the approach, exhibiting a monotonic force output relationship consistent with theory, low hysteresis, high repeatability over repeated cycles, and small response indentation speeds. We further demonstrate integration on a soft robotic gripper, where the module reliably detects grasp events. Compared with liquid filled or wireembedded tactile skins, the proposed modular add on architecture enhances durability, reduces wiring complexity, and supports straightforward deployment across diverse robot geometries. Because the sensing principle reads skin strain patterns, it also suggests extensions to other somatosensory cues such as joint angle or actuator state estimation from surface deformation. Overall, leveraging surface compliance with an external optical module provides a practical and robust route to equip soft robots with force perception while preserving structural flexibility and manufacturability, paving the way for robotic applications and safe human robot collaboration.
Open AccessExternal Photoreflective Tactile Sensing Based on Surface Deformation Measurement
We present a tactile sensing method enabled by the mechanical compliance of soft robots; an externally attachable photoreflective module reads surface deformation of silicone skin to estimate contact force without embedding tactile transducers. Locating the sensor off the contact interface reduces damage risk, preserves softness, and simplifies fabrication and maintenance. We first characterize the optical sensing element and the compliant skin, thendetermine the design of a prototype tactile sensor. Compression experiments validate the approach, exhibiting a monotonic force output relationship consistent with theory, low hysteresis, high repeatability over repeated cycles, and small response indentation speeds. We further demonstrate integration on a soft robotic gripper, where the module reliably detects grasp events. Compared with liquid filled or wireembedded tactile skins, the proposed modular add on architecture enhances durability, reduces wiring complexity, and supports straightforward deployment across diverse robot geometries. Because the sensing principle reads skin strain patterns, it also suggests extensions to other somatosensory cues such as joint angle or actuator state estimation from surface deformation. Overall, leveraging surface compliance with an external optical module provides a practical and robust route to equip soft robots with force perception while preserving structural flexibility and manufacturability, paving the way for robotic applications and safe human robot collaboration.
- Open Access
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 157712-157729
Creative Agents: Simulating the Systems Model of Creativity With Generative Agents
With the growing popularity of generative AI for images, video, and music, we witnessed models rapidly improve in quality and performance. However, not much attention is paid towards enabling AI’s ability to ‘‘be creative’’.We often attribute the quality of ‘‘being creative’’ to an individual or an object, but we believe that countless variables participate in determining what or who is creative, transcending a single entity or artifact. Csikszentmihalyi’s systems model of creativity suggests that creativity is a product of interactions among multiple parts of a society that create, evaluate, and record. In this study, we implemented and simulated Csikszentmihalyi’s systems model of creativity using virtual agents utilizing large language models (LLMs) and text prompts.We conducted experiments in virtual settings where creativity is achieved with the presence of specific characteristics in the artifact. For comparison, the simulations were conducted with two ‘‘virtual artists’’ being 1)in the system, which received feedback from the field, and 2)isolated, which did not. Both agents were compared by analyzing the novelty, which was measured via Creativity Implication Network, and value, quantified through the desired characteristics present in artifacts. Our results suggest that the agents that receive feedback from the field can generate artifacts that are more novel and more valuable, thus more creative, in the framework of the systems model of creativity. Furthermore, the difference becomes more evident when external factors enact changes to the domain.
- 5 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 157712-157729
Open AccessCreative Agents: Simulating the Systems Model of Creativity With Generative Agents
With the growing popularity of generative AI for images, video, and music, we witnessed models rapidly improve in quality and performance. However, not much attention is paid towards enabling AI’s ability to ‘‘be creative’’.We often attribute the quality of ‘‘being creative’’ to an individual or an object, but we believe that countless variables participate in determining what or who is creative, transcending a single entity or artifact. Csikszentmihalyi’s systems model of creativity suggests that creativity is a product of interactions among multiple parts of a society that create, evaluate, and record. In this study, we implemented and simulated Csikszentmihalyi’s systems model of creativity using virtual agents utilizing large language models (LLMs) and text prompts.We conducted experiments in virtual settings where creativity is achieved with the presence of specific characteristics in the artifact. For comparison, the simulations were conducted with two ‘‘virtual artists’’ being 1)in the system, which received feedback from the field, and 2)isolated, which did not. Both agents were compared by analyzing the novelty, which was measured via Creativity Implication Network, and value, quantified through the desired characteristics present in artifacts. Our results suggest that the agents that receive feedback from the field can generate artifacts that are more novel and more valuable, thus more creative, in the framework of the systems model of creativity. Furthermore, the difference becomes more evident when external factors enact changes to the domain.
- Open Access
- 14 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (10) , 10122-10129
TARAD: Task-Aware Robot Affordance-Centric Diffusion Policy Learned From LLM-Generated Demonstrations
In open-ended task settings, the ability of a robot to execute diverse tasks accurately by following language instructions is critical. Methods based on traditional imitation learning typically depend on extensive expert demonstrations and often struggle to generalize in the case of unseen scenarios or tasks. Recently, approaches leveraging large foundational models have demonstrated improved generalization by enhancing task comprehension in novel scenarios based on the intrinsic world knowledge embedded in these models. However, these methods rely on predefined motion primitives and lack a detailed understanding of the environment, which is essential for successful execution. Herein we introduce Task-Aware Robot Affordance-Centric Diffusion Policy (TARAD), a novel framework for robot manipulation. TARAD leverages large language models and vision-language models to perform high-level planning from natural language instructions and extract affordance information from the robot's observations. A heuristic motion planner is employed for low-level motion planning, enabling zero-shot trajectory synthesis and the fully automatic generation of a dataset with language labels and affordances. By incorporating affordances into the observation space, our approach integrates the intrinsic commonsense and reasoning capabilities of foundation models into imitation learning, enabling the training of an affordance-centric, multi-task three-dimensional (3D) diffusion policy. Empirical evaluations in both the RLBench simulated environments and real-world experiments with UR5e demonstrate that TARAD effectively combines the precise control of imitation learning with the strong generalization capabilities of foundation models, all without relying on expert demonstrations or predefined motion primitives.
- 14 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (10) , 10122-10129
Open AccessTARAD: Task-Aware Robot Affordance-Centric Diffusion Policy Learned From LLM-Generated Demonstrations
In open-ended task settings, the ability of a robot to execute diverse tasks accurately by following language instructions is critical. Methods based on traditional imitation learning typically depend on extensive expert demonstrations and often struggle to generalize in the case of unseen scenarios or tasks. Recently, approaches leveraging large foundational models have demonstrated improved generalization by enhancing task comprehension in novel scenarios based on the intrinsic world knowledge embedded in these models. However, these methods rely on predefined motion primitives and lack a detailed understanding of the environment, which is essential for successful execution. Herein we introduce Task-Aware Robot Affordance-Centric Diffusion Policy (TARAD), a novel framework for robot manipulation. TARAD leverages large language models and vision-language models to perform high-level planning from natural language instructions and extract affordance information from the robot's observations. A heuristic motion planner is employed for low-level motion planning, enabling zero-shot trajectory synthesis and the fully automatic generation of a dataset with language labels and affordances. By incorporating affordances into the observation space, our approach integrates the intrinsic commonsense and reasoning capabilities of foundation models into imitation learning, enabling the training of an affordance-centric, multi-task three-dimensional (3D) diffusion policy. Empirical evaluations in both the RLBench simulated environments and real-world experiments with UR5e demonstrate that TARAD effectively combines the precise control of imitation learning with the strong generalization capabilities of foundation models, all without relying on expert demonstrations or predefined motion primitives.