The desired rhythmic signals for adaptive walking of humanoid robots should have proper frequencies, phases, and shapes. Matsuoka's central pattern generator (CPG) is able to generate rhythmic signals with reasonable frequencies and phases, and thus has been widely applied to control the movements of legged robots, such as walking of humanoid robots. However, it is difficult for this kind of CPG to generate rhythmic signals with desired shapes, which limits the adaptability of walking of humanoid robots in various environments. To address this issue, a new framework that can generate desired rhythmic signals for Matsuoka's CPG is presented. The proposed framework includes three main parts. First, feature processing is conducted to transform the Matsuoka's CPG outputs into a normalized limit cycle. Second, by combining the normalized limit cycle with robot feedback as the feature inputs and setting the required learning objective, the neural network (NN) learns to generate desired rhythmic signals. Finally, in order to ensure the continuity of the desired rhythmic signals, signal filtering is applied to the outputs of NN, with the aim of smoothing the discontinuous parts. Numerical experiments on the proposed framework suggest that it can not only generate a variety of rhythmic signals with desired shapes but also preserve the frequency and phase properties of Matsuoka's CPG. In addition, the proposed framework is embedded into a control system for adaptive omnidirectional walking of humanoid robot NAO. Extensive simulation and real experiments on this control system demonstrate that the proposed framework is able to generate desired rhythmic signals for adaptive walking of NAO on fixed and changing inclined surfaces. Furthermore, the comparison studies verify that the proposed framework can significantly improve the adaptability of NAO's walking compared with the other methods.

Recently, the methods based on Convolutional Neural Networks (CNNs) have gained popularity in the field of visual place recognition (VPR). In particular, the features from the middle layers of CNNs are more robust to drastic appearance changes than handcrafted features and high-layer features. Unfortunately, the holistic mid-layer features lack robustness to large viewpoint changes. Here we split the holistic mid-layer features into local features, and propose an adaptive dynamic time warping (DTW) algorithm to align local features from the spatial domain while measuring the distance between two images. This realizes viewpoint-invariant and condition-invariant place recognition. Meanwhile, a local matching DTW (LM-DTW) algorithm is applied to perform image sequence matching based on temporal alignment, which achieves further improvements and ensures linear time complexity. We perform extensive experiments on five representative VPR datasets. The results show that the proposed method significantly improves the CNN-based methods. Moreover, our method outperforms several state-of-the-art methods while maintaining good run-time performance. This work provides a novel way to boost the performance of CNN methods without any re-training for VPR. The code is available at https://github.com/Lu-Feng/STA-VPR .

RGBT tracking is rapidly developing due to its complementary advantages of RGB and thermal frames. Existing methods with high accuracy track at a lower speed, and do not make full use of the hierarchical information in the feature extraction and the historical information of the sequences. To address these issues, a novel dual-modality space-time memory (DMSTM) network is proposed for robust RGBT tracking. Specifically, DMSTM is divided into three modules. The first module is the dual-modality backbone that utilizes both shallow and deep information by aggregating feature maps of dimensional changes during downsampling. Another module is the space-time memory reader with bimodal fusion. It aggregates features of historical and current frames to share information in the time domain. The last module is the siamese head network, which computes the predicted loss sum of the two modalities and back-propagates it. This avoids degrading the tracking performance due to sequence frame pairs where the training targets are not perfectly aligned. Extensive experiments on three RGBT benchmark datasets show that the performance and efficiency of the proposed DMSTM exceed that of state-of-the-art methods while running at 27.6 FPS.

Currently, since the model of a driverless bus is not clear, it is difficult for most traditional path tracking methods to achieve a trade-off between accuracy and stability, especially in the case of driverless buses. In terms of solving this problem, a path-tracking controller based on a Fuzzy Pure Pursuit Control with a Front Axle Reference (FPPC-FAR) is proposed in this paper. Firstly, the reference point of Pure Pursuit is moved from the rear axle to the front axle. It relieves the influence caused by the ignorance of the bus’s lateral dynamic characteristics and improves the stability of Pure Pursuit. Secondly, a fuzzy parameter self-tuning method is applied to improve the accuracy and robustness of the path-tracking controller. Thirdly, a feedback-feedforward control algorithm is devised for velocity control, which enhances the velocity tracking efficiency. The proportional-integral (PI) controller is indicated for feedback control, and the gravity acceleration component in the car’s forward direction is used in feedforward control. Finally, a series of experiments is conducted to illustrate the excellent performances of proposed methods.

Loop closure detection plays a very important role in the mobile robot navigation field. It is useful in achieving accurate navigation in complex environments and reducing the cumulative error of the robot’s pose estimation. The current mainstream methods are based on the visual bag of word model, but traditional image features are sensitive to illumination changes. This paper proposes a loop closure detection algorithm based on multi-scale deep feature fusion, which uses a Convolutional Neural Network (CNN) to extract more advanced and more abstract features. In order to deal with the different sizes of input images and enrich receptive fields of the feature extractor, this paper uses the spatial pyramid pooling (SPP) of multi-scale to fuse the features. In addition, considering the different contributions of each feature to loop closure detection, the paper defines the distinguishability weight of features and uses it in similarity measurement. It reduces the probability of false positives in loop closure detection. The experimental results show that the loop closure detection algorithm based on multi-scale deep feature fusion has higher precision and recall rates and is more robust to illumination changes than the mainstream methods.

Weeds are a significant adverse factor affecting rice growth, and their efficient removal necessitates an accurate, efficient, and well-generalizing weed detection method. However, weed detection faces challenges such as a complex vegetation environment, the similar morphology and color of weeds, and crops and varying lighting conditions. The current research has yet to address these issues adequately. Therefore, we propose GE-YOLO to identify three common types of weeds in rice fields in the Hunan province of China and to validate its generalization performance. GE-YOLO is an improvement based on the YOLOv8 baseline model. It introduces the Neck network with the Gold-YOLO feature aggregation and distribution network to enhance the network’s ability to fuse multi-scale features and detect weeds of different sizes. Additionally, an EMA attention mechanism is used to better learn weed feature representations, while a GIOU loss function provides smoother gradients and reduces computational complexity. Multiple experiments demonstrate that GE-YOLO achieves 93.1% mAP, 90.3% F1 Score, and 85.9 FPS, surpassing almost all mainstream object detection algorithms such as YOLOv8, YOLOv10, and YOLOv11 in terms of detection accuracy and overall performance. Furthermore, the detection results under different lighting conditions consistently maintained a high level above 90% mAP, and under conditions of heavy occlusion, the average mAP for all weed types reached 88.7%. These results indicate that GE-YOLO has excellent detection accuracy and generalization performance, highlighting the potential of GE-YOLO as a valuable tool for enhancing weed management practices in rice cultivation.