Unmanned aerial vehicles (UAVs) have played an important role in recent high-tech local wars. Seizing air control rights with UAVs will undoubtedly be a popular topic future military development. Autonomous combat is complex, antagonistic and mutable, consequently, the decision-making that depends on unmanned systems extremely challenging very little research having been conducted it. An intelligent learning system inspired by mechanisms of brain proposed this paper. In accordance learning,...

In order to quantify the uncertainty of regional wind power generation, this paper proposes a probabilistic forecasting method based on attention mechanism and bidirectional long short-term memory (Bi-LSTM). First, data cleaning is performed historical data, feature set for prediction model training constructed. Then, make selectively focus information with strong correlation, while ignoring weak an introduced capture dependencies relationships between different parts input. Afterwards,...

We need a coordinated control method to continuously track moving target using group of UAVs. In this paper, we study the predicted reference point guidance method, where is considered move with constant velocity in very short time window and trajectories UAVs are designed as several tiny arcs around target. The law UAV divided into roll angle control. Simulations used verify that proposed laws have smaller standoff distances phase errors than Lyapunov vector field model-based predictive...

With the economic development of countries in recent years, influence stock market on global economy has increased. The trends stocks are influenced by many factors, such as information financial results companies, news reports, social media forums, and so on. Stock predictions have also become a popular topic. To date, there been great deal research done professionals, most which used methods related to deep learning. Computer learning Generative Adversarial Networks (GAN) [1],...

Considered the high dynamic and complexity of air combat limitations existing technologies, inspired by knowledge structure human biological neural mechanism learning, this paper designs a brain like to system with autonomous learning ability. Firstly, confrontation ability was divided into action value assessment which belongs category declarative decision-making procedural knowledge. The alternate sub-regional principle using error as motivation designed reference process. simulation...

Guaranteeing the flight with safety and efficiency for UAVs in unknown environment is one of most important problems researchers. In this paper, a novel obstacle avoidance algorithm based on distributional perception decision making proposed, where problem divided into two parts: global making. The optimal path get mission completed safely at last. simulation shows proposed capable solving valuable to improve ability UAVs.

In order to guarantee the adaption of Laguerre Graph algorithm for UAVs path planning, a novel method is proposed with roundness assumption all threat areas and no-fly-zones, after which applied pre-plan flight path. With original shape areas, generated optimized be more suitable UAVs. Simulations show satisfies time veracity requirement.

Humans have a fundamental ability that is to learn others' experience for their own use, while humanoid robots don't have. Several attempts been made specific situations in evolution and study of developmental robots. However, such provided limitations, e.g. learning get overlooked. The present article proposes peers' method, which first reviewed the development as some typical studies revealed, moving from humanlike developmental. These terms are then reconsidered robots' viewpoint,...

Cognitive UAV simulates the intelligence of humans and has ability to sense environment around make decisions by itself. It is more adaptive solve collision avoidance problem in uncertain for UAVs. In this paper, impulsive differential inclusions theory applied establish mathematical model based on characteristic cognitive UAV. On basis, stability defined analyzed form calculated. The simulation shows that analyses will play assistant surveillant role

With the national low-altitude airspace open, small UAVs are widely used in urban space. But there dense obstacles and complex electromagnetic waves, producing serious challenges for UAVs' safety. In region UAV loses track, it should have ability to fly automatically escape from this region. By imitating human-beings' escaping thoughts, five flying principles(exploring principle, optimal path dead zone neighbor gather principle roominess tending principle) proposed based on combination of...

In this paper, we propose a novel tracking method based on structured metric learning, which takes the advantages of both learning and distance learning. our method, is formulated as problem, not only considers importance different samples, but also improves discriminability by specific for matching. Specifically, concrete realized making use constraints from target its neighbour training samples under above framework. Besides, closed-form solution derived problem. To improve matching...

Small UAVs are seeking wide usage in urban space for its advantageous performance. However, there crowded with static and dynamic buildings, causing serious challenges UAVs' safety. This paper proposes a novel autonomous collision avoidance method based on time-obstacle map. First, the state estimation trajectory prediction performed extended Kalman filtering. Second, map is constructed via introducing time axis. Third, flyable paths searched basis of breadth first approach then optimal path...

Collision avoidance is becoming a severe safe challenge for the small UAVs flying among urban buildings. For flight in an unknown area, real-time route planning method needed urgently. This paper just focuses on this problem. Based basic principle of Laguerre diagram, firstly induced single prepared aisle between two And then safest found by analyzing collision error routes. Furthermore, based diagram can be built quad-rotor UAV areas. The presented has been shown that its property fine...

Abstract Speech emotion recognition (SER) is a difficult task because emotions are subjective and recognizing the affective state of speaker challenging. To tackle this issue, Broad Learning System presented to balance training networks that substantially faster than those used previously. Furthermore, we performed experiments on standard IEMOCAP dataset achieved state-of-the-art performance in terms weighted accuracy unweighted accuracy. Taken together, experimental results demonstrated...

This paper analyzes the causes of image noise in seawater and influence on target UUV(unmanned underwater vehicle), points out shortcomings existing methods suppression. In view above problems, we propose a real-time suppression method for UUV platform. The algorithm is divided into three steps: (1) Firstly, binarized by finding an appropriate threshold based dispersion between classes. (2) Then, binary subjected to rapid morphological processing separate sticky noise. (3) Finally, connected...