Searcharxiv⌕ Search

arXiv subjects

Koki Inami

Publications and source records attributed to Koki Inami.

6 recordsLinked to original sources

Autonomous Track Racing with a Commercially Available Humanoid: System Design and Real-World Deployment

Learning-based methods now let humanoids run robustly, and they have begun to compete in athletic events. Yet few reports describe a complete autonomous humanoid running system in an actual race, where the robot must follow a lane at a new venue, pace itself within hardware limits, and finish without intervention. We report such a system deployed on the Unitree G1, a commercially available general-purpose humanoid, in the 100 m, 400 m, and 1500 m events of the World Humanoid Robot Games 2026. The system integrates vision-based lane perception, path tracking, and a running policy trained by reinforcement learning with a human motion prior. To address the challenges of humanoid running, we stabilize head-camera images against body motion and use IMU-assisted leg odometry to bridge flight phases. To manage motor heating, we penalize hip-pitch torque during training, add heat-dissipation plates, and apply temperature-based speed limiting with event-specific pacing. The robot placed 13th of 74 in the 100 m, 10th of 44 in the 400 m first round before falling in the semifinal, and seventh of 35 in the 1500 m. In this deployment, thermal margin set the pace in every event, and the semifinal fall followed a lane-detection error.

cs.RO↗

Reproducible Dynamic Parameter Identification for a Low-Cost Robot Arm: A Positive-Definiteness Audit for Model Acceptance

Dynamic parameter identification of low-cost robot arms is challenging because limited sensing and drivetrain nonidealities can yield models that predict measured torques well but are physically unsuitable for model-based control. This paper presents a reproducible dynamic parameter identification pipeline for CRANE-X7, a low-cost seven-degree-of-freedom arm driven by modular smart actuators. A 39-parameter OpenSYMORO base-parameter model is identified without CAD inertial data using fully specified single-joint and adjacent-pair excitation, ordinary least squares, a conditional semidefinite-programming projection, and closed-loop input error refinement. Model acceptance is decided by a separate positive-definiteness audit of the identified inertia matrix over 221,875 sampled configurations. Experiments cover 40 identification trajectories at four sampling intervals and three held-out validation trajectories. The reduced model improves held-out prediction over the full 65-parameter model on all seven identification trajectories used for the model comparison. Fixed-configuration analyses show the distinct role of the feasibility audit: models with similar torque predictions can produce unstable acceleration-resolved dynamics and reverse the direction of the inertia inversion. For the selected identification trajectory, two executions separated by 26 days yield a 1.25% relative spread in held-out root-mean-square error. The accepted model passes audits with five random seeds, and sensitivity analyses quantify the effects of the sampling interval and of torque-constant uncertainty. The complete trajectory specification and numerical record support independent implementation and comparison. These results establish predictive performance, inertia-matrix feasibility, and repeatability as complementary criteria for evaluatingdynamic models of low-cost arms.

cs.RO↗

Design and Experimental Validation of Sensorless 4-Channel Bilateral Teleoperation for Low-Cost Manipulators

Teleoperation of low-cost manipulators is attracting increasing attention as a practical means of collecting demonstration data for imitation learning. However, most existing low-cost systems rely on unilateral position control without force feedback, while implementing force-feedback bilateral teleoperation is difficult because low-cost manipulators typically have low-resolution encoders and no joint torque sensors. This paper presents a sensorless 4-channel bilateral teleoperation framework that integrates identified nonlinear dynamics compensation with a disturbance-observer-based velocity and external-force estimation scheme. By interpreting the observer structure in the frequency domain, we clarify the coupling between the velocity- and external-force-estimation bandwidths and derive practical tuning guidelines based on the damping ratio and a single cutoff frequency. Real-robot experiments, including force-sensor comparison and teleoperation tasks, demonstrate that the proposed framework provides practically useful force estimates and enables stable teleoperation in high-speed and contact-rich scenarios under low-cost hardware constraints. As an application, imitation-learning experiments demonstrate that incorporating estimated force information into demonstrations improves task success rates in the tested contact-rich manipulation tasks.

cs.RO↗

Motion Generation for Food Topping Challenge 2024: Serving Salmon Roe Bowl and Picking Fried Chicken

Although robots have been introduced in many industries, food production robots are yet to be widely employed because the food industry requires not only delicate movements to handle food but also complex movements that adapt to the environment. Force control is important for handling delicate objects such as food. In addition, achieving complex movements is possible by making robot motions based on human teachings. Four-channel bilateral control is proposed, which enables the simultaneous teaching of position and force information. Moreover, methods have been developed to reproduce motions obtained through human teachings and generate adaptive motions using learning. We demonstrated the effectiveness of these methods for food handling tasks in the Food Topping Challenge at the 2024 IEEE International Conference on Robotics and Automation (ICRA 2024). For the task of serving salmon roe on rice, we achieved the best performance because of the high reproducibility and quick motion of the proposed method. Further, for the task of picking fried chicken, we successfully picked the most pieces of fried chicken among all participating teams. This paper describes the implementation and performance of these methods.

cs.RO↗

Motion ReTouch: Motion Modification Using Four-Channel Bilateral Control

Recent research has demonstrated the usefulness of imitation learning in autonomous robot operation. In particular, teaching using four-channel bilateral control, which can obtain position and force information, has been proven effective. However, control performance that can easily execute high-speed, complex tasks in one go has not yet been achieved. We propose a method called Motion ReTouch, which retroactively modifies motion data obtained using four-channel bilateral control. The proposed method enables modification of not only position but also force information. This was achieved by the combination of multilateral control and motion-copying system. The proposed method was verified in experiments with a real robot, and the success rate of the test tube transfer task was improved, demonstrating the possibility of modification force information.

cs.RO↗

Loss Function Considering Dead Zone for Neural Networks

It is important to reveal the inverse dynamics of manipulators to improve control performance of model-based control. Neural networks (NNs) are promising techniques to represent complicated inverse dynamics while they require a large amount of motion data. However, motion data in dead zones of actuators is not suitable for training models decreasing the number of useful training data. In this study, based on the fact that the manipulator joint does not work irrespective of input torque in dead zones, we propose a new loss function that considers only errors of joints not in dead zones. The proposed method enables to increase in the amount of motion data available for training and the accuracy of the inverse dynamics computation. Experiments on actual equipment using a three-degree-of-freedom (DOF) manipulator showed higher accuracy than conventional methods. We also confirmed and discussed the behavior of the model of the proposed method in dead zones.

cs.RO↗