Exoskeleton robots demonstrate promise in their application in assisting or enhancing human physical capacity. Joint muscular torques (JMT) reflect human effort, which can be applied on an exoskeleton robot to realize an active power-assist function. The estimation of human JMT with a wearable exoskeleton is challenging. This paper proposed a novel human lower limb JMT estimation method based on the inverse dynamics of the human body. The method has two main parts: the inverse dynamic approach (IDA) and the sensing system. We solve the inverse dynamics of each human leg separately to shorten the serial chain and reduce computational complexity, and divide the JMT into the mass-induced one and the foot-contact-force (FCF)-induced one to avoid switching the dynamic equation due to different contact states of the feet. An exoskeleton embedded sensing system is designed to obtain the user’s motion data and FCF required by the IDA by mapping motion information from the exoskeleton to the human body. Compared with the popular electromyography (EMG) and wearable sensor based solutions, electrodes, sensors, and complex wiring on the human body are eliminated to improve wearing convenience. A comparison experiment shows that this method produces close output to a motion analysis system with different subjects in different motion.

Online Continual Learning (OCL), as a core step towards achieving human-level intelligence, aims to incrementally learn and accumulate novel concepts from streaming data that can be seen only once, while alleviating catastrophic forgetting on previously acquired knowledge. Under this mode, the model needs to learn new classes or tasks in an online manner, and the data distribution may change over time. Moreover, task boundaries and identities are not available during training and evaluation. To balance the stability and plasticity of networks, in this work, we propose a replay-based framework for OCL, named Contrastive Correlation Preserving Replay (CCPR), which focuses on not only instances but also correlations between multiple instances. Specifically, besides the previous raw samples, the corresponding representations are stored in the memory and used to construct correlations for the past and the current model. To better capture correlation and higher-order dependencies, we maximize the low bound of mutual information between the past correlation and the current correlation by leveraging contrastive objectives. Furthermore, to improve the performance, we propose a new memory update strategy, which simultaneously encourages the balance and diversity of samples within the memory. With limited memory slots, it allows less redundant and more representative samples for later replay. We conduct extensive evaluations on several popular CL datasets, and experiments show that our method consistently outperforms the state-of-the-art methods and can effectively consolidate knowledge to alleviate forgetting.

A novel balance assistance control strategy of a hip exoskeleton robot was proposed in this paper. The organic fusion of the human balance assessment and the exoskeleton balance assistance control strategy are the assurance of balance recovery. However, currently there are few human balance assessment methods that are suitable for detecting balance loss during standing and walking, and very little research has focused on exoskeleton balance recovery control. In this paper, a single step balance assessment method was proposed first, and then based on this method an "assist-as-needed" balance assistance control strategy was established. Finally, the exoskeleton balance assistance control experiment was carried out. The experiment results verified the effectiveness of the single balance assessment method and the active balance assistance control strategy.

The lower extremity exoskeleton is a device for auxiliary assistance of human movement. The interaction performance between the exoskeleton and the human is determined by the lower extremity exoskeleton’s controller. The performance of the controller is affected by the accuracy of the dynamic equation. Therefore, it is necessary to study the dynamic parameter identification of lower extremity exoskeleton. The existing dynamic parameter identification algorithms for lower extremity exoskeletons are generally based on Least Square (LS). There are some internal drawbacks, such as complicated experimental processes and low identification accuracy. A dynamic parameter identification algorithm based on Particle Swarm Optimization (PSO) with search space defined by Recursive Least Square (RLS) is developed in this investigation. The developed algorithm is named RLS-PSO. By defining the search space of PSO, RLS-PSO not only avoids the convergence of identified parameters to the local minima, but also improves the identification accuracy of exoskeleton dynamic parameters. Under the same experimental conditions, the identification accuracy of RLS-PSO, PSO and LS was quantitatively compared and analyzed. The results demonstrated that the identification accuracy of RLS-PSO is higher than that of LS and PSO.