Accurate and fast human motion pattern recognition is the key to ensuring lower limb assistive devices’ appropriate assistance. The research on human motion pattern recognition of lower limb assistive devices mainly focuses on sagittal gait. The motion pattern such as circular walking (CW) is asymmetric about the sagittal plane of the body. CW is common in daily living. However, the recognition algorithm of CW is rarely reported. Since lower limb assistive devices interact with humans, lacking the capability of recognizing CW is dangerous. Thus, to realize the accurate and fast recognition of CW, this paper proposed a finite class Bayesian interference system (FC-BesIS). FC-BesIS is designed to recognize walking activities (linear walking and CW) and gait events (heel contact, load response, mid stance, terminal stance, pre-swing, initial swing, mid swing, and terminal swing). A finite class method which reduces the number of potential classes according to elimination rules before decision-making is introduced. Elimination rules are designed based on likelihood estimation and sensor information. The experiments show that walking activities and gait events can be accurately and fastly recognized by FC-BesIS. The experiments also show that the performance of FC-BesIS in mean recognition accuracy (MRA) and mean decision time (MDT) is improved compared with BesIS. The MRA of walking activities and gait events are 100% and 97.38%, respectively. The MDT of walking activities and gait events are 28.19 ms and 33.94 ms, respectively. Overall, FC-BesIS has been proved to be an accurate and fast recognition algorithm for human motion patterns using wearable sensors.

The improvement of the steering performance of jet robots is challenging due to single inflexible jet aperture. Scallops provide a potential solution with hard shells and a double-hole jet propulsion, which are expected to achieve fast steering movement under water. Inspired by scallops, a bionic propulsion dynamic mesh is proposed in this article, and a three-dimensional computational model of scallops is established. We further calculated the scallop propulsion mechanism under the swing of shells with different shapes. The coupling of simultaneous swing of two shells and their coupling with velum are presented, revealing the unique movement mechanism of Bivalvia. Based on this, the advantages of the double-hole jet propulsion are applied to develop a scallop robot with excellent steering capabilities. Experiments are conducted to verify the steering performance of the scallop robot.

Objective: To reduce hip joint muscles’ activation during walking with an active hip exoskeleton. Background: Few studies examine the optimal active assistance timing of the hip exoskeleton based on muscle activation characteristics. Methods: Sixteen gender-balanced healthy adults (mean age 28.8 years) performed four tasks (each over 20 min). Tasks were different in loading and assistance. Muscle activation was collected by surface electromyography. The collected oxygen consumption evaluated the performance of the proposed active assistance strategy. Results: Experimental results verified that lower muscle activation and metabolism could be achieved when the active assistance gait phase was 9–60% of the gait cycle than that of all-gait-cycle active assist. Conclusions: Regulating the exoskeleton’s active assistance timing according to muscles’ activation characteristics can improve functional assistance.

Lower-limb exoskeletons require accurate dynamic models to achieve stable and compliant human–robot interactions. However, least-squares-based identification often relies on demanding experiments and may yield limited accuracy for exoskeletons with non-standard structures and actuator-induced uncertainties. This paper proposes a two-stage dynamic parameter identification method that integrates recursive least squares (RLS) and a genetic algorithm (GA), denoted as RLS-GA. RLS is first executed offline to estimate the variation ranges of the inertial parameter vector and to construct a finite, physically meaningful search space. GA then refines the parameters within these bounds by minimizing the regression residual norm. Experiments on a hip exoskeleton show that RLS-GA achieves higher identification accuracy than LS and unconstrained GA, while converging faster than GA under identical conditions.