A Novel Activity Recognition System for Alternative Control Strategies of a Lower Limb Rehabilitation Robot

Abstract

Robot-aided training strategies that allow functional, assist-as-needed, or challenging training have been widely explored. Accurate activity recognition is the basis for implementing alternative training strategies. However, some obstacles to accurate recognition exist. First, scientists do not yet fully understand some rehabilitation activities, such as abnormal gaits and falls; thus, there is no standardized feature for identifying such activities. Second, during the activity identification process, it is difficult to reasonably balance sensitivity and specificity when setting the threshold. Therefore, we proposed a multisensor fusion system and a two-stage activity recognition classifier. This multisensor system integrates explicit information such as kinematics and spatial distribution information along with implicit information such as kinetics and pulse information. Both the explicit and implicit information are analyzed in one discriminant function to obtain a detailed and accurate recognition result. Then, alternative training strategies can be implemented on this basis. Finally, we conducted experiments to verify the feasibility and efficiency of the multisensor fusion system. The experimental results show that the proposed fusion system achieves an accuracy of 99.37%, and the time required to prejudge a fall is approximately 205 milliseconds faster than the response time of single-sensor systems. Moreover, the proposed system also identifies fall directions and abnormal gait types.

New Hybrid AD Methodology for Minimizing the Total Amount of Information Content: A Case Study of Rehabilitation Robot Design

“…detection requirements by arranging the photoelectric sensors at specific spatial locations and applying a novel classifier algorithm. The novel classifier algorithm is detailed in another paper [30][31][32]. Compared to other sensor combination methods, the use of photoelectric sensors dramatically reduces the complexity of the design.…”

Upper Limb Rehabilitation Robot System Based on Internet of Things Remote Control

“…Low, real-time and cost-effective. Use the man-machine interface on the computer (upper computer position) to select different operation modes, and transmit the information to the STM32 controller (lower computer) through serial communication to control the motor to work, so as to drive the mechanical arm to move, and in the same At the time, the position information collected from the encoder and the torque information collected from the torque sensor are fed back to the computer through the STM32 controller [29], which can realize the evaluation, analysis and guidance of patient rehabilitation training [30][31]. As can be seen from Figure 4, the upper limb rehabilitation IoT is the application of parameter information collected by various sensors as parameter variables to the automatically controlled IoT, providing a scientific basis for upper limb rehabilitation robots for data information and video information transmission, and server-side remote control The purpose of client patients for rehabilitation training [32,33].…”

Logic-Enhanced Adaptive Network-Based Fuzzy Classifier for Fall Recognition in Rehabilitation

“…After analyzing the training data, we first set up some simple membership functions (as shown in Fig. 5) and logic rules (as shown in Table 2) based on previous research [28]. The classification results are shown in Fig.…”

“…In [28], a two-stage activity recognition system was proposed. A probabilistic neural network (PNN) is first trained on normal activities, and falls or an abnormal gait are filtered and passed to an SVM-radial basis function-KNN (SVM-RBF-KNN) algorithm; this derives specific falls or abnormal gait from general abnormal activity in the second stage.…”