Zhenbang Gong

Robotic airship mission path-following control based on ANN and human operator’s skill

Robotic airships have numerous low-speed and low-altitude application potentials. Mission path following is one such application, which, however, presents an autonomy challenge. In this paper, a yawing controller, which is based on artificial neural network (ANN) and human operator skills, is proposed for mission path following of robotic airships. First, the path-following errors based on the operator’s point of view are discussed. Then, a data acquisition system is designed to collect the flight data under manual control, and the data are then processed and used for offline training and validation of a multilayer feed-forward neural network. Finally, the trained neural network is reconstructed in the flight control system for yawing control, and the experimental results confirm the effectiveness of this method. It is also shown that the ANN controller is robust even with wind disturbance.

An Attitude Determination System For A Small Unmanned Helicopter Using Low-Cost Sensors

Attitude determination system (ADS) that use inexpensive sensors and are based on computationally efficient and robust algorithms are indispensable for real-time vehicle navigation, guidance and control application. This paper describes an attitude determination system that is based on accelerometer and rate gyro. The algorithm is based on a extended Kalman filter. Using integration of angular rates as state vector and the earths gravity as the measured vector, the attitude (roll and pitch) and gyro bias is estimated. At last, a discrete extended Kalman filter (EKF) implement of the same formulation is test on a small unmanned helicopter.

Real-time Vision-based Object Tracking from a Moving Platform in the Air

Generally the scene of the visual surveillance using one camera is confined within certain limits. If you want to extend the surveillance area, you need multiple cameras. It attempts to detect, recognize and track certain objects from image sequences. This paper presents a simple and effective approach segmenting and tracking a moving object from a moving platform in the air such as UAV or ballonet airship, etc. The moving platform can follow the tracked object during a long distance. It can be used to track the doubtful vehicle and strengthen the visual surveillance more. The proposed approach is efficient and robust. The object is searched only in a subwindow of the current frame. The position of sub-window is estimated by its position in the last two frames. The features of the object are updated in each frame. The approach produces good tracking results at a frame rate of 25 fps. It is available in tracking a moving object in various scenes

Biomimetic control of pan-tilt-zoom camera for visual tracking based-on an autonomous helicopter

A novel control strategy of pan-tilt-zoom camera is described. Because the active camera is mounted on a moving autonomous helicopter in visual tracking system, and the tracked object is moving at same time, and there exists the vibration influence of the helicopter, image stabilization becomes poor, and all pixels are running. Therefore, a biomimetic control strategy of on-board pan-tilt-zoom camera is presented. In this paper, the biomimetic oculomotor control model is obtained based on physiological neural path of eye movement control. In order to validate the functions of the biomimetic control model, simulation experiments were done under the same condition as the physiological experiments in physiological researches. Then the biomimetic controller of onboard pan-tilt-zoom camera is developed. The results of flight tracking experiments show that the biomimetic controller can compensate the deflection caused by the flight platform, and enhance the visual tracking system performance.

Development of cable maintenance robot for cable‐stayed bridges

Purpose – In order to replace the conventional human maintenance of cable‐stayed bridges, a robot is designed and constructed for tasks such as cleaning, painting and rust‐detecting.Design/methodology/approach – Adopting a modular approach, two kinds of climbing mechanisms, plus a painting mechanism and a rust‐detecting method are designed.Findings – A robot that can climb and maintain the cables of cable‐stayed bridges has been designed and constructed. It has been proved by experiment that the robot can overcome many disadvantages of conventional human bridge‐maintenance, and drastically improve efficiency, cost, and safety.Research limitations/implications – The robot is of industrial size, but a new mechanism requiring less installing time will be designed for the future.Practical implications – The robot has been applied to cables of the Nanpu Bridge and Xupu Bridge in Shanghai. More than 80 cable‐stayed bridges and six suspension bridges have been built or are being constructed across large rivers i...

Experimental study on oscillating paddling gait of an eccentric paddle mechanism

Eccentric paddle mechanism (ePaddle) is a novel locomotion mechanism designed for amphibious robot. Integrated with several paddles and a wheel, the ePaddle has versatility in locomotion gaits. In this paper, we focus on the aquatic oscillating paddling gait. The conception of the oscillating paddling gait is introduced firstly and followed by the analysis of the oscillation trajectory of the paddles. In order to verify the ability of producing effective thrust force by the oscillating paddling gait, a thrust measuring facility is built. A series of experiments have been carried out with this facility. From the results, we characterize how the amplitude and frequency of the generated net thrust force relate with the amplitude, period of the oscillation of the paddle. Finally, the influence of the number of paddles on the net thrust is analyzed.

Low Cost Hand Gesture Learning and Recognition System Based on Hidden Markov Model

Focusing on recognizing some typical gestures based on 3-axis MEMS accelerometer to interact with an application of human-machine interactive game, the hand gesture problem is analyzed firstly, then theory basis of HMM is introduced. In order to obtain training gesture data for HMM, and also to provide a hardware basis for gesture recognition, a low cost data acquisition system hardware is researched and designed. The system can get the acceleration data of user’s gesture and transmit them wirelessly to a personal computer. In the hand gesture recognition approach, k-mean algorithms are applied to cluster and abstract the vector data from sensor. And then the quantized vectors are put into a hidden Markov model to learn and recognize user’s gestures. Finally the gesture recognition library is implemented in C# development environment, and is utilized in a human-machine interactive game application. The results show that the typical gesture emerging in the game can be identified in a high rate, and the user can experience more interest and interaction.

A Pan-tilt Camera Control System of UAV Visual Tracking Based on Biomimetic Eye

Firstly, the characteristics of UAV visual tracking system were analysed, secondly the three-dimensional oculomotor control system model of biomimetic eye was developed based on the oculomotor neural paths. Then a new UAV visual tracking system was established. The on-board pan-tilt camera control system based on biomimetic eye can compensate the deflection caused by the UAV rotation and the movement of ground relative to the UAV. The pan-tilt camera can simulate human eye movements including smooth pursuit, VOR and OKR, and the control pattern of camera motor is velocity control, thus it can track the ground moving target continuously, smoothly, clearly and stably like human eye. Finally the experimental results were presented.

Non-reciprocating legged gait for robot with epicyclic-gear-based eccentric paddle mechanism

A novel eccentric paddle mechanism based on epicyclic gear mechanism (ePaddle-EGM) has been proposed to enhance the mobility of amphibious robots for multi-terrain tasks with diverse locomotion gaits. This paper proposes a unique non-reciprocating legged walking gait for a quadruped or hexapod robot based on ePaddle-EGMs after a brief kinematic analysis of the ePaddle-EGM. During the gait, the robot can generate desired periodic motion of the legs in the swinging and supporting phases while maintaining all the actuators rotating unidirectionally. Advantages of the proposed non-reciprocating legged gaits include that influence of backlash in the epicyclic gear mechanism could be eliminated and accuracy of locomotion can be guaranteed. Experiments on a single ePaddle-EGM prototype module were conducted to validate the proposed mechanism design and the idea of deploying the non-reciprocating legged walking gait to improve locomotion accuracy and energetic efficiency of the robot. A non-reciprocating legged gait is proposed for an ePaddle-EGM-based robot.Gait sequence and planning method for this gait are discussed and verified.Locomotion accuracy is secured with the gait in the presence of gear backlash.Robot walks more efficiently in this gait than in legged crawling gait.

Experimental study on the oscillating paddling gait of an ePaddle mechanism with flexible configuration

In this paper, a novel eccentric paddle locomotion mechanism (ePaddle) has been proposed to enhance the mobility of amphibious robots for multi-terrains tasks with diverse locomotion gaits. The oscillating paddling gait of the ePaddle mechanism enables the robot to perform stationary observation or attitude maneuvering operations in shallow water. To increase the thrust generated by this gait, the ePaddle mechanism has a flexible configuration, i.e. a flexible paddle and three rigid paddles. The effects of the oscillating amplitude and period of the gait to thrust are analyzed and compared with the thrusts measured with rigid configuration. Experimental results demonstrate that the flexible configuration is able to produce much more net thrust than the rigid configuration when the ePaddle is oscillating at large amplitude. Graphical Abstract