An Improved Extreme Learning Machine (ELM) Algorithm for Intent Recognition of Transfemoral Amputees With Powered Knee Prosthesis.

To overcome the challenges posed by the complex structure and large parameter requirements of existing classification models, the authors propose an improved extreme learning machine (ELM) classifier for human locomotion intent recognition in this study, resulting in enhanced classification accuracy. The structure of the ELM algorithm is enhanced using the logistic regression (LR) algorithm, significantly reducing the number of hidden layer nodes. Hence, this algorithm can be adopted for real-time human locomotion intent recognition on portable devices with only 234 parameters to store. Additionally, a hybrid grey wolf optimization and slime mould algorithm (GWO-SMA) is proposed to optimize the hidden layer bias of the improved ELM classifier. Numerical results demonstrate that the proposed model successfully recognizes nine daily motion modes including low-, mid-, and fast-speed level ground walking, ramp ascent/descent, sit/stand, and stair ascent/descent. Specifically, it achieves 96.75% accuracy with 5-fold cross-validation while maintaining a real-time prediction time of only 2 ms. These promising findings highlight the potential of onboard real-time recognition of continuous locomotion modes based on our model for the high-level control of powered knee prostheses.

Design, Manufacture, and Experimental Validation of a Hydraulic Semi-Active Knee Prosthesis.

In this article, a new hydraulic semi-active knee (HSAK) prosthesis is proposed. Compared with knee prostheses driven by hydraulic-mechanical coupling or electromechanical systems, we novelly combine independent active and passive hydraulic subsystems to solve the incompatibility between low passive friction and high transmission ratio of current semi-active knees. The HSAK not only has the low friction to follow the intentions of users, but also performs adequate torque output. Moreover, the rotary damping valve is meticulously designed to effectively control motion damping. The experimental results demonstrate the HSAK combines the advantages of both passive and active prostheses, including the flexibility of passive prostheses, as well as the stability and the sufficient active torque of active prostheses. The maximum flexion angle in level walking is about 60°, and the peak output torque in stair ascent is greater than 60Nm. Relative to the daily use of prosthetics, the HSAK improves gait symmetry on the affected side and contributes to the amputees better maintain daily activities.

Theoretical and Experimental Investigations into a Crawling Robot Propelled by Piezoelectric Material.

Conventional motors with complicated electromagnetic structures are difficult to miniaturise for millimetre- and centimetre-sized robots. Instead, small-scale robots are actuated using a variety of functional materials. We proposed a novel robot propelled by a piezoelectric ceramic in this work. The robot advances due to the asymmetric friction created by the spikes on the surface. The structural modelling was completed, static and dynamic models were established to predict the moving characteristics, the prototype was built using three dimensional (3D) printing technology, and the models were evaluated via experiments. Compared with conventional inchworm-type robots, the proposed robot is superior in simple structure because the clamping components are replaced by spikes with asymmetric friction. Compared with SMA (shape memory alloy) actuating inchworm-type robots, it has a faster velocity with higher resolution. Meanwhile, the components are printed through an additive manufacturing process that is convenient and avoids assembly errors. This design could make contributions to many areas, such as pipe inspection, earthquake rescue, and medicine delivery.

The Impact of Locomotor Speed on the Human Metatarsophalangeal Joint Kinematics.

This paper aims to further our previous study to investigate the effect of speed on the human metatarsophalangeal (MP) joint kinematics during running on level ground. The 3D motion of the foot segments was captured by a twelve-camera motion analysis system, and the ground reaction forces and moments were recorded by using a six-force plate array. The relative movement between the tarsometatarsi (hindfoot) and phalanges (forefoot) segments were recorded to obtain the 3D orientation and position of the functional axis (FA) of the MP joint. The results show that the FA locates about an average of 19% foot length (FL) anterior to the anatomical axis (AA) across all running speeds, and is also 4.8% FL inferior to the AA during normal and fast run. Similar to walking, the functional axis is more oblique than the anatomical axis with a more anterior-inferior orientation across all the running speeds. This suggests that representing MP joint with the AA may mislead the calculation of joint moment/power and muscle moment arms in both running and walking gait. Compared with previous study, we found that walking and running speeds have statistically significant effects on the position of the FA. The functional axis moves frontward to a more anterior position when the speed increases during walking and running. It transfers upward in the superior direction with increasing speed of walking, but moves more toward the inferior position when the velocity increased further to running. Also, the orientation of FA in sagittal plane became more oblique toward the vertical direction as the speed increased. This may help in moderating the muscular effort, increase the muscle EMA and improve the locomotor performance. These results would contribute to understanding the in vivo biomechanical function of the MP joint and also the foot propulsion during human locomotion.