This paper proposes a long-term forecasting scheme and implementation method based on the interval type-2 fuzzy sets theory for traffic flow data. The type-2 fuzzy sets have advantages in modeling uncertainties becaus...
详细信息
This letter proposes a robust stochastic differential equation approach for learning point-to-point motions in an adversarial way. The proposed stochastic dynamical model combines the advantages of the stochastic diff...
详细信息
This letter proposes a robust stochastic differential equation approach for learning point-to-point motions in an adversarial way. The proposed stochastic dynamical model combines the advantages of the stochastic differential equation and the transformer-like function together to achieve both robustness and accuracy of the learning. The adversarial training method is proposed to simplify the way of updating the parameters of the model. The state of the proposed stochastic dynamical system is mathematically proved to converge asymptotically in the mean square sense, and it has been experimentally validated on the LASA dataset and by the trajectory-programming task of the Franka Emika robot. The experimental results show that: (1) the adversarial training method helps the model to achieve higher reproduction accuracy;(2) the trajectories generated by the proposed model achieve higher accuracy in both the noise-free condition (by approximately 14.9%) and the noisy condition (by approximately 17.8%) compared with the state-of-the-art methods in terms of the similarity to the demonstration;and (3) the proposed approach can learn smoother trajectories even if the observations are contaminated by noises.
This paper focuses on the gaits planning method of the backward swimming for unsymmetrical structure bio-inspired robotic fish. Based on the differences between the anguilliform mode and carangiform mode swimming, a m...
详细信息
This paper focuses on the gaits planning method of the backward swimming for unsymmetrical structure bio-inspired robotic fish. Based on the differences between the anguilliform mode and carangiform mode swimming, a method for searching gaits of backward swimming was proposed to plan the motion of the developed carangiform robotic fish. The body envelope of European eel's backward swimming was mimicked according to the freely swimming model, which was proposed to analyze the propulsion produced by the undulation of the multi-link tail. Finally, simulations and experiments were conducted to demonstrate the gaits searching method for the bio-inspired carangiform robotic fish.
In this paper, we discuss a novel graph matching problem, namely the parameterized Koopmans- Beckmann's graph matching (KBGMw). KBGMw is defined by a weighted linear combination of a series of Koopmans-Beckmann...
详细信息
In this paper, we discuss a novel graph matching problem, namely the parameterized Koopmans- Beckmann's graph matching (KBGMw). KBGMw is defined by a weighted linear combination of a series of Koopmans-Beckmann's graph matching. First, we show that KBGMw can be taken as a special case of the parameterized Lawler's graph matching, subject to certain conditions. Second, based on structured SVM, we propose a supervised learning method for automatically estimating the parameters of KBGMw. Experimental results on both synthetic and real image matching data sets show that the proposed method achieves relatively better performances, even superior to some deep learning methods. (c) 2020 Elsevier B.V. All rights reserved.
Text information contained in scene images is very helpful for high-level image understanding. In this study, the authors propose to learn co-occurrence of local strokes for scene text recognition by using a spatialit...
详细信息
Text information contained in scene images is very helpful for high-level image understanding. In this study, the authors propose to learn co-occurrence of local strokes for scene text recognition by using a spatiality embedded dictionary (SED). Unlike spatial pyramid partitioning images into grids to incorporate spatial information, the authors SED associates every codeword with a particular response region and introduces more precise spatial information for robust character recognition. After localised soft coding and max pooling of the first layer, a sparse dictionary is learned to model co-occurrence of several local strokes, which further improves classification performance. Experimental results on two scene character recognition datasets ICDAR2003 and CHARS74 K demonstrate that their character recognition method outperforms state-of-the-art methods. Besides, competitive word recognition results are also reported for four benchmark word recognition datasets ICDAR2003, ICDAR2011, ICDAR2013 and street view text when combining their character recognition method with a conditional random field language model.
Soft pressure sensors have recently attracted considerable attention because of their applications in human-machine interface, soft robotics, and prosthetics. However, there remain some challenges in achieving satisfa...
详细信息
Soft pressure sensors have recently attracted considerable attention because of their applications in human-machine interface, soft robotics, and prosthetics. However, there remain some challenges in achieving satisfactory performance (e.g., high sensitivity, wide sensing range, high stability) for soft pressure sensors. This article reports an intentional blocking based photoelectric pressure sensor. Two different blocking methods are investigated: the single-row-pyramid blocking and the double-row-pyramid blocking. The sensor has a simple structure, which is made of a light-emitting diode, photosensitive element, and silicone sensor shell. Experiments demonstrate that the sensor has a high sensitivity (the maximum sensitivity is 48.07 kPa(-1), and the minimum measurement pressure is 0.8 Pa), large pressure-sensing range (the sensing range is up to 120 kPa), superior stability (a drift about 0.4% over 12,130 repetitive cycles at 0-80 kPa), low drift (< +/- 0.2% in different 3-day testing), negligible hysteresis, and high signal-to-noise ratio (over 55 dB). By mounting the pressure sensor at the end of a robotic arm, the robot can detect subtle collisions (such as touching a balloon through a pinpoint). In addition, this article fabricates a tactile glove based on the proposed pressure sensor and shows the application of this glove for music playing and object weighing. This study provides a new structure for photoelectric sensors to increase sensitivity and also provides a more convenient way to fabricate photoelectric pressure sensors.
This paper addresses the problem of three-dimensional (3-D) waypoint tracking for a biomimetic underwater vehicle (BUV) propelled by undulatory fins: RobCutt-II. Based on the specific mechanical design and control sys...
详细信息
This paper addresses the problem of three-dimensional (3-D) waypoint tracking for a biomimetic underwater vehicle (BUV) propelled by undulatory fins: RobCutt-II. Based on the specific mechanical design and control system configuration, the RobCutt-II can perform diversified locomotion patterns, especially submerging or surfacing vertically in the water. For practical underwater operating procedures, a selective switching control for 3-D waypoint tracking is proposed. This control scheme contains a depth controller, a waypoint tracking controller, and a selector. When tracking a series of given 3-D waypoints, the RobCutt-II can switch between two closed-loop locomotion patterns, i.e., the depth control pattern and the waypoint tracking pattern. Simulations and a comparative experimental study demonstrate the feasibility and effectiveness of the proposed switching control scheme.
Most conventional microscopes can only acquire two-dimensional (2D) images. However, there is increasing demand for microscopes capable of acquiring three-dimensional (3D) images for use in biological research. To imp...
详细信息
Most conventional microscopes can only acquire two-dimensional (2D) images. However, there is increasing demand for microscopes capable of acquiring three-dimensional (3D) images for use in biological research. To improve observations of in vivo specimens, we develop a novel one snap-shot 3D light field microscope with high spatial resolution. By adding a lenslet array and focusing optics into the optical train of a conventional microscope, we are able to simultaneously capture seven sub-aperture images from different perspectives in a single shot, and finish 3D surface reconstruction of the specimen. In this paper, we present a functioning prototype of a novel 3D light field microscope using different objectives, and describe its design and image quality. During 3D image reconstruction, we use an optical flow method and back-projection interpolation to improve the resolution of the subimages. The performance of the novel light field microscope is analyzed theoretically and experimentally. Multiple perspective images are obtained using an objective with 20 x /NA 0.4. These images are 1050 x 1050 pixels with a depth of view of 200 mu m. For our microscope, the highest resolvable group of USAF 1951 target is about group 7.5. 3D reconstructions of various specimens are successfully created using these images. This novel microscopy system is suitable for various applications in 3D imaging and optical metrology. (C) 2017 Elsevier B.V. All rights reserved.
In an environment where robots coexist with humans, mobile robots should be human-aware and comply with humans' behavioural norms so as to not disturb humans' personal space and activities. In this work, we pr...
详细信息
In an environment where robots coexist with humans, mobile robots should be human-aware and comply with humans' behavioural norms so as to not disturb humans' personal space and activities. In this work, we propose an inverse reinforcement learning-based time-dependent A* planner for human-aware robot navigation with local vision. In this method, the planning process of time-dependent A* is regarded as a Markov decision process and the cost function of the time-dependent A* is learned using the inverse reinforcement learning via capturing humans' demonstration trajectories. With this method, a robot can plan a path that complies with humans' behaviour patterns and the robot's kinematics. When constructing feature vectors of the cost function, considering the local vision characteristics, we propose a visual coverage feature for enabling robots to learn from how humans move in a limited visual field. The effectiveness of the proposed method has been validated by experiments in real-world scenarios: using this approach robots can effectively mimic human motion patterns when avoiding pedestrians;furthermore, in a limited visual field, robots can learn to choose a path that enables them to have the larger visual coverage which shows a better navigation performance.
In this paper, a multiple-object tracking approach in large-scale scene is proposed based on visual sensor network. Firstly, the object detection is carried out by extracting the HOG features. Then, object tracking is...
详细信息
In this paper, a multiple-object tracking approach in large-scale scene is proposed based on visual sensor network. Firstly, the object detection is carried out by extracting the HOG features. Then, object tracking is performed based on an improved particle filter method. On the one hand, a kind of temporal and spatial dynamic model is designed to improve the tracking precision. On the other hand, the cumulative error generated from evaluating particles is eliminated through an appearance model. In addition, losses of the tracking will be incurred for several reasons, such as occlusion, scene switching and leaving. When the object is in the scene under monitoring by visual sensor network again, object tracking will continue through object re-identification. Finally, continuous multiple-object tracking in large-scale scene is implemented. A database is established by collecting data through the visual sensor network. Then the performances of object tracking and object re-identification are tested. The effectiveness of the proposed multiple-object tracking approach is verified.
暂无评论