Visual Simultaneous Localization and Mapping (SLAM) in dynamic scenes is a prerequisite for robot-related applications. Most of the existing SLAM algorithms mainly focus on dynamic object rejection, which makes part of the valuable information lost and prone to failure in complex environments. This paper proposes a semantic visual SLAM system that incorporates rigid object tracking. A robust scene perception frame is designed, which gives autonomous robots the ability to perceive scenes similar to human cognition. Specifically, we propose a two-stage mask revision method to generate fine mask of the object. Based on the revised mask, we propose a semantic and geometric constraint (SAG) strategy, which provides a fast and robust way to perceive dynamic rigid objects. Then, the motion tracking of rigid objects is integrated into the SLAM pipeline, and a novel bundle adjustment is constructed to optimize camera localization and object' 6-DoF poses. Finally, the evaluation of the proposed algorithm is performed on publicly available KITTI dataset, Oxford Multimotion Dataset, and real-world scenarios. The proposed algorithm achieves the comprehensive performance of RPEt less than 0.07m per frame and RPER about 0.03° per frame in the KITTI dataset. The experimental results reveal that the proposed algorithm enables accurate localization and robust tracking than state-of-the-art SLAM algorithms in challenging dynamic scenarios.
Abstract In this paper, trajectory tracking control is investigated for lower extremity rehabilitation exoskeleton robot. Unknown perturbations are considered in the system which are inevitable in the reality. The trajectory tracking control is constructively treated as constrained control issue. To obtain the explicit equation of motion and analytical solution of lower extremity rehabilitation exoskeleton robot, Udwadia-Kalaba theory is introduced. Lagrange multipliers and pseudo variables are not needed in Udwadia-Kalaba theory, which is more superior than Lagrange method. On the basic of Udwadia-Kalaba theory, two constrained control methods including trajectory stabilization control and adaptive robust control are proposed. Trajectory stabilization control applies Lyapunov stability theory to modify the desired trajectory constraint equations. A leakage-type of adaptive law is designed to compensate unknown perturbations in adaptive robust control. Finally, comparing with nominal control and control method in [32], simulation results demonstrate the superiority of trajectory stabilization control and adaptive robust control in trajectory tracking control.
A new adaptive control scheme based on neural network is proposed. It combines neural network with optimum method to solve the control law. It is shown by simulation that using this algorithm, the system responses fast and has small overshoot without steady error. The simulation results have verified the effectiveness of these algorithms.
Abstract Real-time detection of conveyor belt tearing is of great significance to ensure mining in the coal industry. The longitudinal tear damage problem of conveyor belts has the characteristics of multi-scale, abundant small targets, and complex interference sources. Therefore, in order to improve the performance of small-size tear damage detection algorithms under complex interference, we propose a visual detection method YOLO-STOD based on deep learning. Firstly, a multi-case conveyor belt tear datasets is developed for complex interference and small-size detection. Second, the detection method YOLO-STOD is designed, which utilizes the BotNet attention mechanism to extract multi-dimensional tearing features, enhancing the model's feature extraction ability for small targets and enables the model to converge quickly under the conditions of few samples. Secondly, Shape_IOU is utilized to calculate the training loss, and the shape regression loss of the bounding box itself is considered to enhance the robustness of the model. Finally, the detection performance of the designed algorithm in complex environments is verified. The experimental results show that the proposed algorithm has high detection accuracy and detection rate compared with existing detection algorithms, and it is expected to be used for real-time detection of conveyor belt tearing in the industrial field.
Visual navigation based on deep reinforcement learning requires a large amount of interaction with the environment, and due to the reward sparsity, it requires a large amount of training time and computational resources. In this paper, we focus on sample efficiency and navigation performance and propose a framework for visual navigation based on multiple self-supervised auxiliary tasks. Specifically, we present an LSTM-based dynamics model and an attention-based image-reconstruction model as auxiliary tasks. These self-supervised auxiliary tasks enable agents to learn navigation strategies directly from the original high-dimensional images without relying on ResNet features by constructing latent representation learning. Experimental results show that without manually designed features and prior demonstrations, our method significantly improves the training efficiency and outperforms the baseline algorithms on the simulator and real-world image datasets.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)