As a key technology for the non-invasive human-machine interface that has received much attention in the industry and academia, surface EMG (sEMG) signals display great potential and advantages in the field of human-machine collaboration. Currently, gesture recognition based on sEMG signals suffers from inadequate feature extraction, difficulty in distinguishing similar gestures, and low accuracy of multi-gesture recognition. To solve these problems a new sEMG gesture recognition network called Multi-stream Convolutional Block Attention Module-Gate Recurrent Unit (MCBAM-GRU) is proposed, which is based on sEMG signals. The network is a multi-stream attention network formed by embedding a GRU module based on CBAM. Fusing sEMG and ACC signals further improves the accuracy of gesture action recognition. The experimental results show that the proposed method obtains excellent performance on dataset collected in this paper with the recognition accuracies of 94.1%, achieving advanced performance with accuracy of 89.7% on the Ninapro DB1 dataset. The system has high accuracy in classifying 52 kinds of different gestures, and the delay is less than 300 ms, showing excellent performance in terms of real-time human-computer interaction and flexibility of manipulator control.

There are a variety of objects, random postures and multiple objects stacked in a disorganized manner in unstructured home applications, which leads to the object grasping posture estimation and grasping planning based on machine vision become very complicated. This paper proposes a method of cluttering pose detection based on convolutional neural network with multiple self-powered sensors information. Firstly, a search strategy for candidate grasping poses based on the 3D point cloud is proposed, and the single-channel grasping image dataset representing grasping posture in this paper is established by using Bigbird dataset. Secondly, the ResNet is constructed to rank and filter the single channel captured images representing the captured bit pose. It is also compared with three mainstream classification networks, Inception V2, VGG-A and LetNet, and the perception analysis function and the execution planning function are developed under ROS. The effective grasping of the manipulator in the scene of scattered piles is realized based on the detection results of grasping position and combined with the information of multiple self-powered sensors, and the ResNet network is compared with other three classification networks. In a scattered and stacked environment of objects, the results of experiment show that the method based on ResNet network is superior to the other three networks, and the average success rate of grasping pose detection based on ResNet, InceptionV2, VGGA and LetNet networks is 90.67%, 82.67%, 86.67% and 87.33% respectively, which verifies the effectiveness and superiority of the deep learn-based grasping pose detection model proposed in this paper.

Due to the increased employment of robots in modern society, path planning methods based on human–robot collaborative mobile robots have been the subject of research in both academia and industry. The dynamic window approach used in the research of the robot local path planning problem involves a mixture of fixed weight coefficients, which makes it hard to deal with the changing dynamic environment and the issue of the sub-optimal global planning paths that arise after local obstacle avoidance. By dynamically modifying the combination of weight coefficients, we propose, in this research, the use of fuzzy control logic to optimize the evaluation function’s sub-functions and enhance the algorithm’s performance through the safe and dynamic avoidance of obstacles. The global path is introduced to enhance the dynamic window technique’s ability to plan globally, and important points on the global path are selected as key sub-target sites for the local motion planning phase of the dynamic window technique. The motion position changes after local obstacle avoidance to keep the mobile robot on the intended global path. According to the simulation results, the enhanced dynamic window algorithm cuts planning time and path length by 16% and 5%, respectively, while maintaining good obstacle avoidance and considering a better global path in the face of various dynamic environments. It is difficult to achieve a local optimum using this algorithm.

With the rapid development of artificial intelligence, neural network is widely used in various fields. Target detection algorithm is mainly based on neural network, but the accuracy of target detection algorithm is greatly related to the complexity of scene and texture. A target detection algorithm based on RGB-D image from the perspective of the lightweight of target detection network model and the integration of depth map to overcome the weak environmental illumination with self-powered sensors information is proposed. This paper analyzes the network model structure of YOLOv4 and MobileNet, compares the variation of parameter number between depthwise separable convolution and convolutional neural network, and combines the advantages of YOLOv4 network and MobileNetv3 network. The main network of three effective feature layers in YOLOv4 is replaced by MobileNetv3 network for initial feature layer extraction to strengthen the feature extraction network. At the same time, the standard convolution models in the network are replaced by depthwise separable convolution. The proposed method is compared with YOLOv4 and YOLOv4-MobileNetv3 in this paper, and the experimental results show that the proposed network retains its original advantages in accuracy, but the size of the network model is about 23% of that of YOLOv4 network model, and the processing speed is about 42% higher than that of YOLOv4 network model, and the detection accuracy can still reach 83% in the environment with poor lighting conditions.

Autonomous driving has gradually become a research hotspot in recent years, but the robustness of loopback detection in complex environments such as dynamic and weak textures needs to be improved. A semantic loopback detection method is proposed based on instance segmentation and visual SLAM to make sufficient use of semantic information in autonomous driving. The proposed method combines image segmentation and visual SLAM (Simultaneous Localization and Mapping) to construct a semantic SLAM system. What’s more, a data association method that combines semantic and geometric information is proposed to improve the traditional loopback detection method by using semantic information to increase the accuracy of loopback detection. The result of experiment on the TUM public dataset shows that the loopback detection accuracy of the improved loopback detection method is higher than that of the bag-of-words method in all four datasets, and our proposed algorithm can effectively improve the accuracy of loopback detection of the SLAM system in general.

In dynamic environments, several simultaneous localization and mapping (SLAM) systems effectively utilize optical flow fields to distinguish dynamic from static feature points. Commonly, these systems leverage the amplitude information within the optical flow field to develop adaptive thresholding segmentation models for identifying dynamic scene regions. Nevertheless, designing adaptive thresholding models typically necessitates meticulous planning and extensive experimentation. This study introduces a dynamic RGBD SLAM system, SFD-SLAM, which innovates by employing a saliency detection network for the direct extraction of dynamic regions via scene flow. This approach notably streamlines the design process associated with conventional adaptive thresholding models. Furthermore, SFD-SLAM incorporates a geometric module that merges depth residuals with hyperpixel segmentation to enhance the refinement of the dynamic mask. This is followed by integration with FCM clustering for the precise identification of moving objects. The efficacy of SFD-SLAM is assessed using the widely recognized TUM dynamic dataset. Experimental results demonstrate that the proposed system surpasses DGFlow-SLAM, which relies on an adaptive thresholding model for dynamic object segmentation, in terms of trajectory accuracy. It also achieves comparable localization accuracy to DynaSLAM. Moreover, SFD-SLAM maintains robust tracking capabilities, even in scenarios where DynaSLAM experiences tracking loss, thereby augmenting the robustness of RGBD-SLAM in dynamic settings.