No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A local target enhanced reconnaissance algorithm based on autonomous exploration of robots
ResearchGate has not been able to resolve any citations for this publication.
The article proposes a real‐time apple picking method based on an improved YOLOv5. This method accurately recognizes different apple targets on fruit trees for robots and helps them adjust their position to avoid obstructions during fruit picking. Firstly, the original BottleneckCSP module in the YOLOv5 backbone network is enhanced to extract deeper features from images while maintaining lightweight. Secondly, the ECA module is embedded into the improved backbone network to better extract features of different apple targets. Lastly, the initial anchor frame size of the network is adjusted to avoid recognizing apples in distant planting rows. The results demonstrate that this improved model achieves high accuracy rates and recall rates for recognizing various types of apple picking methods with an average recognition time of 0.025s per image. Compared with other models tested on six types of apple picking methods, including the original YOLOv5 model as well as YOLOv3 and EfficientDet‐D0 algorithms, our improved model shows significant improvements in mAP by 1.95%, 17.6%, and 12.7% respectively. This method provides technical support for robots' picking hands to actively avoid obstructions caused by branches during fruit harvesting, effectively reducing apple loss.
Landslides are the most common type of geological disaster in China, causing substantial property losses and casualties. It is of great significance for disaster prevention and mitigation to quickly and accurately identify landslide areas. Currently, the use of faster region-based convolutional neural networks (Faster R-CNN) for landslide detection is very effective. However, the deep feature extraction capabilities of this network may lead to gradient vanishing during the training process, making it impossible to adjust the network layer weights. Therefore, this paper proposes a new remote-sensing image object detection algorithm based on the Faster R-CNN algorithm. In this paper, the main feature extraction networks, VGG16 and ResNet50, in the Faster R-CNN algorithm are replaced by DarkNet53, which uses skip connections to overcome the problem of gradient disappearance caused by excessively deep networks. Then, a data augmentation method is applied to the landslide dataset to solve the problem of the insufficient number of landslide samples in the training set. Finally, experiments are conducted to compare the proposed algorithm with Faster R-CNN (VGG16) and Faster R-CNN (ResNet50) and to evaluate the effectiveness of the proposed algorithm. The results show that the AP value of our algorithm can reach 92.05%, the F1 score can reach 0.80, the recall value can reach 93.80%, and the precision value can reach 69.59% on the same landslide dataset. Compared to the Faster R-CNN (VGG16) algorithm, the AP value increased by 10.71%, and compared to the Faster R-CNN (ResNet50) algorithm, the AP value increased by 0.42%.
In this paper, we propose a scheme for autonomous exploration in unknown environments using a mobile robot. To reduce the storage consumption and speed up the search of frontiers, we propose a wave-features-based rapidly exploring random tree method, which can inhibit or promote the growth of sampling trees regionally. Then, we prune the frontiers with mean shift algorithm and use the pruned frontiers for decision-making. To avoid the repeated exploration, we develop a decision making method with consistency assessment, in which the status of the robot and frontiers are explicitly encoded and modeled as a fixed start open traveling salesman problem (FSOTSP). Furthermore, a re-decision mechanism is build to reduce the extra computing cost. Simulations and real-world experiments show the significant improvement of the proposed scheme.
Recently, deep learning-based classification approaches have made great progress and now dominate a wide range of applications, thanks to their Herculean discriminative feature learning ability. Despite their success, for hyperspectral data analysis, these deep learning based techniques tend to suffer computationally as the magnitude of the data soars. This is mainly because the hyperspectral imagery (HSI) data are multidimensional, as well as giving equal importance to the large amount of temporal and spatial information in the HSI data, despite the redundancy of information in the temporal and spatial domains. Consequently, in literature, this equal information emphasis has proven to affect the classification efficacy negatively in addition to increasing the computational time. As a result, this paper proposes a novel dual branch spatial-spectral attention based classification methodology that is computationally cheap and capable of selectively accentuating cardinal spatial and spectral features while suppressing less useful ones. The theory of feature extraction with 3D-convolutions alongside a gated mechanism for feature weighting using bi-directional long short-term memory is used as a spectral attention mechanism in this architecture. In addition, a union of 3D convolutional neural network (3D-CNN) and a residual network oriented spatial window-based attention mechanism is proposed in this work. To validate the efficacy of our proposed technique, the features collected from these spatial and spectral attention pipelines are transferred to a feed-forward neural network (FNN) for supervised pixel-wise classification of HSI data. The suggested spatial-spectral attention based hyperspectral data analysis and image classification methodology outperform other spatial-only, spectral-only, and spatial-spectral feature extraction based hyperspectral image classification methodologies when compared, according to experimental results.
Because traditional methods are difficult to solve the problems related to the path planning of logistics robots, this study proposes a method of using computer multimedia 3D reconstruction technology to realize the path planning of warehouse logistics robots. Without fully considering the accurate movement path between points, according to the warehouse logistics robot, it is judged whether the starting point is at the exit. The planning problem of the movement path is converted into a TSP problem and a TS-TSP problem. Finally, the analysis of experimental results shows that the method proposed in this study converges faster than traditional algorithms and can quickly obtain the global optimal solution. At the same time, the warehousing logistics robot requires less path planning time and has strong practical application.
We present a method for autonomous exploration in complex three-dimensional (3D) environments. Our method demonstrates exploration faster than the current state-of-the-art using a hierarchical framework-one level maintains data densely and computes a detailed path within a local planning horizon, while another level maintains data sparsely and computes a coarse path at the global scale. Such a framework shares the insight that detailed processing is most effective close to the robot, and gains computational speed by trading-off details far away from the robot. The method optimizes an overall exploration path with respect to the length of the path and produces a kinodynamically feasible local path. In experiments, our systems autonomously explore indoor and outdoor environments at a high degree of complexity, with ground and aerial robots. The method produces 80% more exploration efficiency, defined as the average explored volume per second through a run, and consumes less than 50% of computation compared to the state-of-the-art.
A great number of visual simultaneous localization and mapping (VSLAM) systems need to assume static features in the environment. However, moving objects can vastly impair the performance of a VSLAM system which relies on the static-world assumption. To cope with this challenging topic, a real-time and robust VSLAM system based on ORB-SLAM2 for dynamic environments was proposed. To reduce the influence of dynamic content, we incorporate the deep-learning-based object detection method in the visual odometry, then the dynamic object probability model is added to raise the efficiency of object detection deep neural network and enhance the real-time performance of our system. Experiment with both on the TUM and KITTI benchmark dataset, as well as in a real-world environment, the results clarify that our method can significantly reduce the tracking error or drift, enhance the robustness, accuracy and stability of the VSLAM system in dynamic scenes.
Automatic and fast weld type classification, tacked spot recognition and weld ROI (region of interest) determination are key links for intelligent welding robot, because they directly affect the seam tracking and welding parameters such as current, voltage and torch inclination. Nevertheless, there are few studies on weld type classification, tacked spot recognition, and weld ROI determination. Considering the fast inference speed of YOLOv5, an automatic weld type classification, tacked spot recognition and weld ROI determination based on modified YOLOv5 is presented in this paper. First, the detection requirements of weld type classification, tacked spot recognition and weld ROI determination are transformed into a unified target localization task to improve the inference speed, so the three results can be obtained through single inference; the next, to improve the localization accuracy of weld ROI, the center component bias between the predicted box and ground truth is added to the original CIOU localization loss function; then, a weighted classification loss function is used to reduce the false positives in fillet and groove welds; finally, a self-template method for padding image border is presented to improve the generalization ability of the trained model. Experimental results show that: the presented method reaches 100% precision, 100% recall, 0.91 mean intersection-over-unio, 2.41 pixels center component bias of determined weld ROI and 18 ms inference times in the original size images; the center component bias of determined weld ROI is reduced from 2.38 pixels to 2.18 pixels by adding the center component bias loss to CIOU function in the padded images; the weighted classification loss function reduced the false positives in fillet and groove welds; compared with the default gray border padding method, the model trained by using self-template padding method reduced the center component bias of the determined weld ROI from 2.73 pixels to 2.41 pixels in the original size images. Moreover, when Ref. Chen et al. (2022) positions the welding seam coordinates in the weld ROI determined by the presented method, the recall is improved from 0.96 to 0.98, and the computation time is reduced from 180 ms to 48 ms.
Object detection techniques are the foundation for the artificial intelligence field. This research paper gives a brief overview of the You Only Look Once (YOLO) algorithm and its subsequent advanced versions. Through the analysis, we reach many remarks and insightful results. The results show the differences and similarities among the YOLO versions and between YOLO and Convolutional Neural Networks (CNNs). The central insight is the YOLO algorithm improvement is still ongoing.This article briefly describes the development process of the YOLO algorithm, summarizes the methods of target recognition and feature selection, and provides literature support for the targeted picture news and feature extraction in the financial and other fields. Besides, this paper contributes a lot to YOLO and other object detection literature.
In this paper, we propose a novel path planning framework for autonomous exploration in unknown environments using a mobile robot. A graph structure is incrementally constructed along with the exploration process. The structure is the road map that represents the topology of the explored environment. To construct the road map, we design a sampling strategy to get random points in the explored environment uniformly. A global path from the current location of the robot to the target area can be found on this road map efficiently. We utilize a lazy collision checking method that only checks the feasibility of the generated global path to improve the planning efficiency. The feasible global path is further optimized with our proposed trajectory optimization method considering the motion constraints of the robot. This mechanism can facilitate the path cost evaluation for the next best view selection. In order to select the next best target region, we propose a utility function that takes into account both the path cost and the information gain of a candidate target region. Moreover, we present a target reselection mechanism to evaluate the target region and reduce the extra path cost. The efficiency and effectiveness of our approach are demonstrated using a mobile robot in both simulation and real experimental studies.
Reconnaissance and autonomy for small robots (RASR) team: MAGIC 2010 challenge
- A Lacaze
- K Murphy
- M Giorno
- K Corley