Action assessment and feedback can effectively assist fitness practitioners in improving exercise benefits. In this paper, we address key challenges in human action recognition and assessment by proposing innovative methods that enhance performance while reducing computational complexity. Firstly, we present Oct-MobileNet, a lightweight backbone network, to overcome the limitations of the traditional OpenPose algorithm’s VGG19 network, which exhibits a large parameter size and high device requirements. Oct-MobileNet employs octave convolution and attention mechanisms to improve the extraction of high-frequency features from the human body contour, resulting in enhanced accuracy with reduced model computational burden. Furthermore, we introduce a novel approach for action recognition that combines skeleton-based information and multiple feature fusion. By extracting spatial geometric and temporal characteristics from actions, we employ a sliding window algorithm to integrate these features. Experimental results show the effectiveness of our approach, demonstrating its ability to accurately recognize and classify various human actions. Additionally, we address the evaluation of traditional fitness exercises, specifically focusing on the BaDunJin movements. We propose a multimodal information-based assessment method that combines pose detection and keypoint analysis. Label sequences are obtained through a pose detector and each frame’s keypoint coordinates are represented as pose vectors. Leveraging multimodal information, including label sequences and pose vectors, we explore action similarity and perform quantitative evaluations to help exercisers assess the quality of their exercise performance.

In order to address the fragility issues associated with the current prescribed performance control (PPC) strategy and ensure both transient and steady-state performance of the tracking error, a non-fragility prescribed performance control scheme is proposed. A non-fragile prescribed performance control method for robotic systems with model uncertainties and unknown disturbances is developed. This method not only addresses the inherent vulnerability defects of the existing prescribed performance control but also effectively reduces the computational complexity of the controller. Firstly, addressing the fragility issues of existing PPC, a new non-fragile prescribed performance control strategy is proposed. To address the fragile issue with the current PPC, the shift function is employed to handle the tracking error. Based on the non-fragile PPC mentioned above, a new prescribed performance controller is designed without the requirement for approximation or estimation. This effectively reduces the complexity of controller design. At last, the feasibility of achieving non-fragile prescribed performance is verified through stability analysis, and the superiority of the designed controller is confirmed through simulation comparisons. The results show that the designed controller effectively resolves the control singularity issue arising from the inherent limitations of the PPC.

visual odometry (VO) systems typically rely on stacked convolutional layers and Long Short-Term Memory (LSTM) units to capture long-range dependencies within sequences. However, convolutional neural networks are relatively weak in modeling global context information, making them ill-suited to addressing drift issues that arise during long-term navigation. Furthermore, their inherent locality impedes the model from achieving better performance. Therefore, we propose a transformer-based visual odometry method, named TSCVO, to address these challenges. TSCVO consists of two main components: 1) A relative pose subnetwork based on the k-NN Timeformer model, which utilizes divided space-time attention to estimate relative poses between adjacent frames. It substitutes k-NN attention for the original fully connected self-attention in the Timeformer model to ignore irrelevant token information and reduce computational complexity. 2) A global pose subnetwork based on cross-frame attention, which utilizes cross-frame interaction mechanisms to learn long-range dependencies and internal correlations within image sequences. Additionally, we introduce robust A − Loss as the foundational loss function, supplemented by geometric consistency loss to further constrain structural similarity between adjacent frames and improve prediction accuracy. The evaluation results on the KITTI and EuRoC datasets indicate that TSCVO outperforms other learning-based methods.