Search

Wind perception is crucial for autonomous sailboat which has features of long distance operation. Previously, anemometers based on thermal dynamics, acoustics or cup-vane anemometer have already been widely used, but they are generally large in size. This paper starts with vibration mechanics to develop a new kind of anemometer which is particularly suitable for mini-sized autonomous sailboats. An IMU which is fixed to one end of a non-rigid body, such as spring. The non-rigid body will vibrate if it is exposed to wind field. Physical quantities related to vibration can reflect different wind speeds. We modeled the new design and developed a prototype with a low cost. Experiments are conducted to collect data. Based on curve fitting of the experiment data, the wind speed is estimated with a mean error at 0.35 m/s for measuring range from 0 to 10.7 m/s.

In this paper, we propose a navigation algorithm oriented to multi-agent environment. This algorithm is expressed as a hierarchical framework that contains a Hidden Markov Model (HMM) and a Deep Reinforcement Learning (DRL) structure. For simplification, we term our method Hierarchical Navigation Reinforcement Network (HNRN). In high-level architecture, we train an HMM to evaluate the agents perception to obtain a score. According to this score, adaptive control action will be chosen. While in low-level architecture, two sub-systems are introduced, one is a differential target-driven system, which aims at heading to the target; the other is a collision avoidance DRL system, which is used for avoiding dynamic obstacles. The advantage of this hierarchical structure is decoupling the target-driven and collision avoidance tasks, leading to a faster and more stable model to be trained. The experiments indicate that our algorithm has higher learning efficiency and rate of success than traditional Velocity Obstacle (VO) algorithms or hybrid DRL method.

Personal handwriting style fonts generation is a diverting but time-consuming task due to the large size of Chinese character set. In addition, unlike standard printed style fonts, hand-writing style fonts are of more complicated stroke and glyph feature. In this paper, an improved network architecture is proposed for learning and generation of personal hand-writing style fonts based on small character set. The network is composed of three sub-networks: 1) a classification network for identifying the general style of the target fonts; 2) a generating network for transferring the identified fonts to the target fonts; 3) a discriminating network for differentiating the generated image from real ones. The experiments revealed the effectiveness of the model for generating personal hand-writing style font with relatively small data size, reduction by a scale of 10 comparing to previous reported works.

This paper presents a new model of automatic music generation which is based on simulated annealing and genetic algorithm. Our model uses simulated annealing algorithm to create melody units on different chords, and regards them as the genes in genetic algorithm, which can pick out a fine combination of these genes. This model is efficient for producing massive fair-sounding pieces of double-track music, when provided with a customized metric. Hence, it can be especially useful for massively producing background musics for video games, or providing inspirations for music composers. In this paper, we detailedly state how to transform the prior knowledge about music theory into a reasonable metric, and propose a sample metric for reference. Using this sample metric, we generate a set of music pieces as test cases. The evaluation results show that these pieces fully satisfy the need of quality and creativity.

The ability to maintain balance and stability is a major challenge for quadruped robots in locomotion. For example, when walking in the unstructured environment, the robot not only needs to adjust its gait according to the terrain information, but also needs to tune the body posture based on the real-time body posture acquired from inertial measurement unit (IMU) to maintain stability. In this paper, we present a method that focuses on the design of the posture correction controller based on the IMU feedback for adaptive slope walking for a quadruped robot. To this end, the posture correction controller is integrated into the foot trajectory generator to maintain balance. We experimentally validated our approach both on simulation and real quadruped robot. Compared with the one only with the foot trajectory generator, the quadruped robot can walk more stable with the posture correction controller integrated into the foot trajectory generator and has a better locomotion ability. Our method increases locomotion stability when walking on the slopes.

In this paper, we present an accurate approach to estimate vehicles' pose and shape from off-board multiview images. The images are taken by monocular cameras and have small overlaps. We utilize state-of-the-art convolutional neural networks (CNNs) to extract vehicles' semantic keypoints and introduce a Cross Projection Optimization (CPO) method to estimate the 3D pose. During the iterative CPO process, an adaptive shape adjustment method named Hierarchical Wireframe Constraint (HWC) is implemented to estimate the shape. Our approach is evaluated under both simulated and real-world scenes for performance verification. It's shown that our algorithm outperforms other existing monocular and stereo methods for vehicles' pose and shape estimation. This approach provides a new and robust solution for off-board visual vehicle localization and tracking, which can be applied to massive surveillance camera networks for intelligent transportation., Comment: 8 pages, 8 figures, published in ROBIO 2018

Trajectory tracking is a decisive factor that affects autonomous driving performance of mobile robots in real-world applications, such as industry, service and other areas. Nowadays, there has been many works about path planning and trajectory tracking. Whereas there are problems to be improved to make mobile robot well applied in practical world. For example, generated trajectory can't be tracked smoothly, and the robot has to stop and rotate itself to adjust pose and pass segments in trajectory because of constrains of path curvature and robot's kinematics model. In this paper, a new trajectory planning algorithm with cubic spline interpolation method was proposed, which fully considered constrains of robot model including velocity and acceleration bounds. And corresponding trajectory was generated. Then a robust controller was built to conduct tracking tests. Experimental results demonstrate that our planning and tracking algorithm could help robot achieve a smooth, accurate and efficient trajectory tracking performance without any stop.

In the field of autonomous driving, abundant researches have been carried out in development of advanced Lidar or vision algorithms for mapping and localization, and some efforts in the field of sensor fusion have been made to combine the advantages of multiple sensor types. However, most works focused on the 2D map construction, which had its own limitation with complex environments. In this paper, a new algorithm based on the fusion of Lidar and Inertial Measurement Unit (IMU) data is developed to construct a 2.5D map. We integrate the height information transformed from IMU sensor with laser scan results to construct a 2.5D occupied grid map. The proposed algorithm is convenient to implement on real-world mobile vehicles, and it is fast and robust. The effectiveness of our algorithm is validated by both simulation and practical experiment.

The challenges in transforming paintings into music is well-known, since the relationship between two kinds of art is unclear. Different composers write different music when the same painting is presented to them. In this paper, a cross-mordality model has been proposed for transforming images into multitrack music based on the framework of deep convolutional generative adversarial networks (DCGANs). The proposed model is trained on a classical music dataset and a dataset of impressionist paintings. The model can be applied to transfer impressionist paintings into classical music with two tracks. By using music evaluation methods, the harmonicity of the generated music can be confirmed. Our model is the first attempt of our knowledge at transforming paintings into music segments.

Leader-follower system could expand the load capacity in transport tasks. While few related literatures pay attention to the physical limitation and limit the maximum output velocity of designed controller. This paper expands a trajectory tracking controller with velocity constraints to leader-follower system. Leader and follower are both subjected to linear velocity constraint and angular velocity constraint due to the physical limitation. Selected trajectory tracking controller is applied in leader-follower system. Meanwhile, noises in reference trajectory of follower are released by Kalman filter. Then, a physical experiment platform with Turtlebots is established to verify this control scheme. Finally, experiment results of several examples proved the practicability of proposed control system. The velocity commands of this control system would not exceed the limitation of physical limitation.