KYKenjiro Yamamoto

Kenjiro Yamamoto

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Kenjiro Yamamoto has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Modality Attention for Prediction-Based Robot Motion Generation: Improving Interpretability and Robustness of Using Multi-Modality

    by Hideyuki Ichiwara, Hiroshi Ito, Kenjiro Yamamoto, Hiroki Mori, Tetsuya Ogata

    We developed a modality attention motion generation model on the basis of multi-modality prediction. This model provides interpretability about modality usage and demonstrates robustness against disturbances. We used a hierarchical model consisting of low-level recurrent neural networks (RNNs) for processing each modality individually and a high-level RNN that integrates the multi-modality. This integration is achieved by efficiently gating multi-modality and inputting it to the high-level RNN. We verified the interpretability and robustness of the task of inserting a furniture part, which consists of the “approach” phase to bring the wooden dowel closer to the hole and the “insertion” phase. While the proposed model achieves the same task success rate as the conventional model, it clarifies that it refers to vision during “approach” and force during “insertion,” providing interpretability regarding modality use. Furthermore, in contrast to the non-modality attention model, whose task success rate drops significantly under disturbance, the proposed model enhances robustness against disturbances to modalities it does not direct attention during the task, resulting in a consistently high success rate ( $\simeq 90\%$ ).


  • Open Access

    Real-time Motion Generation and Data Augmentation for Grasping Moving Objects with Dynamic Speed and Position Changes

    by Kenjiro Yamamoto, Hiroshi Ito, Hideyuki Ichiwara, Hiroki Mori, Tetsuya Ogata

    While deep learning enables real robots to perform complex tasks had been difficult to implement in the past, the challenge is the enormous amount of trial-and-error and motion teaching in a real environment. The manipulation of moving objects, due to their dynamic properties, requires learning a wide range of factors such as the object's position, movement speed, and grasping timing. We propose a data augmentation method for enabling a robot to grasp moving objects with different speeds and grasping timings at low cost. Specifically, the robot is taught to grasp an object moving at low speed using teleoperation, and multiple data with different speeds and grasping timings are generated by down-sampling and padding the robot sensor data in the time-series direction. By learning multiple sensor data in a time series, the robot can generate motions while adjusting the grasping timing for unlearned movement speeds and sudden speed changes. We have shown using a real robot that this data augmentation method facilitates learning the relationship between object position and velocity and enables the robot to perform robust grasping motions for unlearned positions and objects with dynamically changing positions and velocities.

  • Efficient multitask learning with an embodied predictive model for door opening and entry with whole-body control

    by Hiroshi Ito, Kenjiro Yamamoto, Hiroki Mori, Tetsuya Ogata

    Robots need robust models to effectively perform tasks that humans do on a daily basis. These models often require substantial developmental costs to maintain because they need to be adjusted and adapted over time. Deep reinforcement learning is a powerful approach for acquiring complex real-world models because there is no need for a human to design the model manually. Furthermore, a robot can establish new motions and optimal trajectories that may not have been considered by a human. However, the cost of learning is an issue because it requires a huge amount of trial and error in the real world. Here, we report a method for realizing complicated tasks in the real world with low design and teaching costs based on the principle of prediction error minimization. We devised a module integration method by introducing a mechanism that switches modules based on the prediction error of multiple modules. The robot generates appropriate motions according to the door’s position, color, and pattern with a low teaching cost. We also show that by calculating the prediction error of each module in real time, it is possible to execute a sequence of tasks (opening door outward and passing through) by linking multiple modules and responding to sudden changes in the situation and operating procedures. The experimental results show that the method is effective at enabling a robot to operate autonomously in the real world in response to changes in the environment.


  • Two-View Monocular Depth Estimation by Optic-Flow-Weighted Fusion

    by Alex Masuo Kaneko, Kenjiro Yamamoto

    Depth estimation with monocular cameras is a cheap and promising solution for autonomous vehicles and robots. Even though there are many approaches in literature, the issue of estimating depth of objects with low optic flow (low parallax) still remains. This work proposes a new two-view monocular depth estimation method that estimates depths with only a monocular camera using two optic flow directions based on the Flat Surface Model, fusing them with optic flow as weights. The proposed method achieved an average depth estimation error of 3.68 m and maximum error of 107.34 m, which was smaller than the obtained by traditional techniques (22.90 and 9815.44 m).


  • Generation of Peg Insert Motions by a Recurrent Neural Network Using Motor Joint Angles and Current Values

    by Takumi Kurata, Hiroshi Ito, Hiroki Mori, Kenjiro Yamamoto, Tetsuya Ogata

    In this research, we propose a motion generation method for inserting rod-shaped pegs from the neighborhood of the hole by using only the joint angles and current values measured from the robot arm’s motors. Conventionally, expensive torque sensors are used about peg-in-hole. However, we confirmed that the peg inserting motion can be generated by learning a neural network using not torque sensors but the motor angles and current values. The learning model is a recursive neural Network integrated and learned by time series data composed of joint angles and current values made with a remote teaching system and it predicts the next motion at a certain time by its position and variance. We confirmed that flexible peg insertion can be realized from multiple initial positions by feedback control with proposed method using motor's joint angles and current values at each time after learning.

  • Basic Design of Human-Symbiotic Robot EMIEW

    by Yuji Hosoda, Saku Egawa, Junichi Tamamoto, Kenjiro Yamamoto, Ryousuke Nakamura, Masahito Togami

    We are developing a robot that supports people in their daily lives: a human-symbiotic robot. Such robot must share space with its users, be user-friendly, and be able to assist its users. We have developed a prototype autonomous mobile robot that makes use of a self-balancing two-wheeled mobile system and a body swing mechanism to shift its center of gravity. This allows it to move nimbly at up to 6 km per hour. It also has capabilities to avoid collisions with obstacles for moving safely through complex environments. Distant-speech-recognition and high-quality speech-synthesis technologies enable it to communicate with people naturally (i.e., without special tools). These capabilities were demonstrated at the 2005 World Exposition in Aichi, Japan