Recent advances in research have demonstrated that Vision-Language Models (VLMs) are a promising technology for robot task planning. This paper presents a novel approach that leverages visual prompts and VLMs to generate feasible robot action sequences for achieving shared tasks through human-robot collaboration while simultaneously estimating human intentions. Our method enhances VLMs’ understanding of the environment by utilizing annotations (bounding boxes and labels) and dynamically infers human intentions based on changing environmental conditions to generate optimal robot action sequences to achieve common goals. Additionally, the system incorporates a mechanism to regenerate new sequences through VLM analysis when action failures or external interference occur. Furthermore, by designing prompts as versatile modules for diverse tasks, our proposed technology offers a new approach to robot action planning that excels in both efficiency and adaptability.

This paper presents the bit error rate (BER) performance for various quadrature amplitude modulation (QAM) schemes with low-density parity-check (LDPC) coding using pilot symbol assisted (PSA) - extended Kalman filter (EKF) phase-noise compensation (PNC) in the presence of strong phase noise (PN) for orthogonal frequency division multiplexing (OFDM) waveforms in sub-terahertz (sub-THz) bands. In the sub-THz bands, PN is a significant hardware impairment and comprises random Wiener PN and uncorrelated Gaussian PN. Hence, in this paper, we propose a compensation method for the correlated Wiener PN by using PSA-EKF PNC and to make use of the large coding gain from the LDPC code to mitigate the effect of the uncorrelated Gaussian PN in the sub-THz bands. Based on computer simulations, we show that the loss in the required received signal-to-noise ratio (SNR) at the BER of $10^{-6}$ compared to the case without PN is suppressed to less than 1.0 dB for both QPSK and 16QAM with the LDPC coding rate of $R=8 / 9$ for the power spectral density level of an uncorrelated Gaussian PN of $-105 \mathrm{dBc} / \mathrm{Hz}$ or lower. We also show that the 64QAM scheme achieves the BER of $10^{-6}$ by utilizing the LDPC coding gain at the cost of an increase in the required received SNR in the presence of strong PN in the sub-THz bands.

Soft robotic hands are reliable for grasping objects of various shapes. However, they perform poorly in the high-precision manipulation of grasped objects because the modeling and sensing of soft actuator deformation are complex. To overcome this problem, in a previous study, we proposed a robust visual feedback control method for precise in-hand manipulation using parallel soft actuators. This method enables precise in-hand manipulation without measuring the soft actuator deformations. Generally, in the feedback control of a parallel drive system, the actuator force is converted into the force/torque applied to the grasped object. The conversion matrix constantly changes depending on the position/orientation of the object and the contact points of the soft actuators with the object. Consequently, accurately measuring the conversion matrix is difficult. Therefore, we estimated it as a constant matrix in our previous studies, and its robustness was confirmed experimentally. However, its theoretical robustness was not analyzed sufficiently. Therefore, in this article, we discuss the robustness of the estimated constant matrix through a mathematical stability proof. Furthermore, we investigate the characteristics of the estimated drive matrix. Then, we perform a numerical robustness analysis. Finally, the robustness of the proposed method is studied via the above investigation and verification experiments.

A pneumatically-driven flexible finger equipped with a flexible sensor is realized for improving the performance of the soft robotic hand. First, we propose a flexible angle estimation sensor. This sensor measures the change in the amount of light passing through polyurethane material and estimates the angle with high repeatability. Next, we design a flexible finger that makes this sensor easy to incorporate. The flexible fingers are produced with a multi-material 3D printer that can use flexible material. The flexible finger can accommodate the proposed flexible sensor within it. It is possible to place the sensor's signal line in the air pressure pipeline. Because the flexible finger is produced with a 3D printer, variations in each model's characteristics are small as compared with manufacturing through molding. In this paper, we show an improvement of positional accuracy in the proposed flexible finger using angle feedback control from the proposed sensor. The effectiveness of this sensor is also shown to solve the problem of vibration problems for the flexible finger during high speed motion.