Human knee joint has a yaw-axis rotational DOF and a locking mechanism called screw-home mechanism(SHM). We focus on this mechanism and implement it to a musculoskeletal humanoid through hardware design. In this paper, we first explained human knee characteristic of SHM and joint range of motion. Next, we explained hardware design and implementation method of the mechanism. As an evaluation of our developed knee joint, we checked the function of SHM from viewpoint of moment arm of the yaw rotational axis and the yaw angle displacement during squat motion. Lastly, as unique and integrated motions that involve the use of yaw DOF derived from SHM, we made several motion achievement including humanlike twisting squat, heel move motion and switch pedals, under a condition of contacting environment. This result demonstrates human mimetic rotational DOF of knee contributes motion achievement.
RT and ICT devices such as robots for assisting daily living are beginning to be introduced to architectural spaces where humans are present. However, there is concern that some metropolitan areas are suffering electromagnetic environmental effects that overwhelm the electromagnetic susceptibility of many electronic devices. RT and ICT devices must function in accordance with the performance characteristics of buildings and electrical facilities within spaces where many electronic devices operate. Accordingly, to enable people to achieve a high quality of life and safety by using RT and ICT devices, we need to build spaces where such technologies can function reliably. In this study, we discuss this issue and present electromagnetic environmental criteria for utilizing spaces where electronic devices function. We also present technologies for measuring and evaluating electromagnetic environmental effects and implementing necessary countermeasures.
We developed a prototype thin plate type tactile sensor for detecting surface shapes such as micro-step shapes and rough shapes. A piezoelectric polymer (polyvinylidene fluoride; PVDF), which has stress (strain) rate dependent output, is glued on the thin plate. Therefore, when the tip of this thin plate scans target uneven surfaces, sharp changes on the surfaces are easily sensed. In this study, we showed that this simple sensing system experimentally detects a single micro-step shape clearly up to a scanning speed of 20[mm/s], which is higher than that of the popular stylus type surface roughness tester. The characteristics of the experimental output signals agree with the estimated signals. Moreover, we showed that this tactile sensor can detect different triangle wave-shape surface roughness (arithmetic average roughness Ra: 1.6–6.3[μm]) limited to a low scanning speed; this subtle shape detection was impossible by using a strain gauge as the sensor element instead of PVDF.