Simultaneous Localization and Mapping (SLAM) is one of the fundamental capabilities for intelligent mobile robots to perform state estimation in unknown environments. However, most visual SLAM systems rely on the static scene assumption and consequently have severely reduced accuracy and robustness in dynamic scenes. Moreover, the metric maps constructed by many systems lack semantic information, so the robots cannot understand their surroundings at a human cognitive level. In this paper, we propose SG-SLAM, which is a real-time RGB-D semantic visual SLAM system based on the ORB-SLAM2 framework. First, SG-SLAM adds two new parallel threads: an object detecting thread to obtain 2D semantic information and a semantic mapping thread. Then, a fast dynamic feature rejection algorithm combining semantic and geometric information is added to the tracking thread. Finally, they are published to the ROS system for visualization after generating 3D point clouds and 3D semantic objects in the semantic mapping thread. We performed an experimental evaluation on the TUM dataset, the Bonn dataset, and the OpenLORIS-Scene dataset. The results show that SG-SLAM is not only one of the most real-time, accurate, and robust systems in dynamic scenes, but also allows the creation of intuitive semantic metric maps.

In the context of the problem of water pollution, the movement characteristics and patterns of fish under normal water quality and abnormal water quality are clearly different. This paper proposes a biological water quality monitoring method combining three-dimensional motion trajectory synthesis and integrated learning. The videos of the fish movement are captured by two cameras, and the Kuhn-Munkres (KM) algorithm is used to match the target points of the fish body. The Kalman filter is used to update the current state and find the optimal tracking position as the tracking result. The Kernelized Correlation Filters (KCF) algorithm compensates the targets that are lost in the tracking process and collision or occlusion in the movement process, reducing the errors caused by illumination, occlusion and water surface fluctuation effectively. This algorithm can directly obtain the target motion trajectory, avoiding the re-extraction from the centroid point in the image sequence, which greatly improves the efficiency. In order to avoid the one-sidedness of the two-dimensional trajectory, the experiment combines the pixel coordinates of different perspectives into three-dimensional trajectory pixel coordinates, so as to provide a more authentic fish swimming trajectory. We then select a representative positive and negative sample data set; the number of data sets should have symmetry. The base classifier capable of identifying different water quality is obtained by training. Finally, support vector machine(SVM), eXtreme Gradient Boosting (XGBoost) and pointnet based classifiers are combined into strong classifiers through integrated learning. The experimental results show that the integrated learning model can reflect the water quality effectively and accurately under the three-dimensional trajectory pixel coordinates of fish, and the recognition rate of water quality is above 95%.

Human waist action recognition based on surface EMG (Electromyography) signals is a complex pattern recognition problem, the difficulty of which lies in the coupling of EMG signals between different actions, and a single model cannot effectively utilize the feature information to achieve high accuracy of action recognition. To this end, this paper proposes a Stacking integration-based EMG signal model for the waist, which introduces the Stacking integration framework into EMG signal recognition for the first time, using KNN (K-Nearest Neighbor), RF (Random Forest-master), XGboost, and LightGBM algorithms as base learners and DT (Decision tree) model as a meta-learner to construct the integrated model. The complex feature information contained in the EMG signal is correlated with different waist movements to construct feature datasets with high correlation, and the training set is divided by using the five-fold cross-validation method to reduce the overfitting problem during the repeated training of the integrated model. Using the integrated model proposed in this paper, the generalization ability of the model can be effectively improved and the recognition accuracy of EMG signals can be improved. According to experiments, the model has an average accuracy of 94.98% for the recognition of six waist movements in normal person, and the accuracy of recognition of the six movements in patients with lumbar strain was 91.5% on average, which is about 6% better than a single model, and the recognition speed can be within 150ms, meeting the requirements of real-time recognition.

Active compliance control represents a pivotal technique for managing the interplay between forces and positions. This article proposes an adaptive approach based on an identified model for variable trajectory control. Addressing parameter redundancy in the dynamic model, we propose a simplified, friction-incorporated modeling method. Key parameters are identified through current and torque data, with three friction models designed to approximate real-world scenarios. A method for monitoring instability risks is presented, particularly effective when the mass difference between the manipulator and the mobile platform is relatively small. It monitors imprudent power output, reducing risks such as overturning. Finally, within system safety constraints, an impedance model is designed to manage the relationship between external forces and position in Cartesian space. This model extends the range of motion to counteract external force influence, utilizing a trajectory modulation technique. Additionally, a trajectory tracking controller based on super-twisting sliding mode control is designed, and the stability of the control system is analyzed. Experimental validation confirms the efficacy of the proposed methodology.

The lumbar EMG(Electromyography) can effectively reflect the current activity state of human lumbar muscles; however, cardiac signals in the EMG significantly impacts the accuracy of lumbar EMG analysis. Currently, there are some problems in the research of EMG signal interference removal, such as poor removal effects and low operation efficiencies, which do not meet the requirement of real-time signal processing. To solve this problem, this paper proposes a method for removing ECG(Electrocardiogram) interference from lumbar EMG signals based on signal segmentation and SSA(Singular Spectrum Analysis), which first, reduces the amount of data for SSA by detecting and segmenting the ECG interference signal segments and then separating the ECG signals present in the EMG signals through four steps of embedding, decomposition, grouping and reconstruction using SSA, effectively removing the ECG signals present in the EMG signals. Finally, a complete lumbar surface EMG signal without ECG interference was obtained by using the recombination method. The method not only provides a better removal effect but can also greatly improve the computing efficiency of the algorithm. It was proven that the method can improve the computing efficiency by 50-80%, and for signals with obvious ECG interference, the computing speed of the algorithm can be increased to the millisecond level, thus achieving real-time processing.

Increasing demand for mobile manipulators in fields requiring high-precision tasks has introduced new requirements regarding their kinematic accuracy. Pose measurement of the mobile manipulator is a crucial method to improve the absolute positioning accuracy of the robot. To solve the pose estimation of the mobile manipulator, we propose a space measurement method based on visual marks and an interactive filtering method to improve the stability of measurement results. Firstly, the kinematics and dynamics of the mobile manipulator are analyzed, and the mathematical expression of the model is obtained to obtain the theoretical trajectory. Secondly, a visual marker based on pyramid hierarchy is proposed, which can adapt to scenes with significant distance scale changes. A visual marker of a regular 12-face is designed, the marker of each face is uniquely coded to ensure that multiple markers can be captured at the same time under a single viewing angle, and geometric relations determine the distance and pose of the end. We propose a filtering method considering the correlation of historical data to improve the stability and accuracy of measurement data. The matching degree of the filter motion mode is improved through the Pearson coefficient relationship between the filtered data in the current period and the historical data. In this paper, several experiments verify the method’s effectiveness. The repeatability of the measurement can reach 0.75mm, and the mean error(MAV)is 0.009mm. The noise test proves the anti-interference ability of the filtering method. The new filtering method has excellent performance on various tracking tracks. All experiments show that the method proposed in this paper has a good measurement effect and is a low-cost, convenient, and efficient method to solve the pose estimation and precise trajectory error analysis of mobile manipulators.