Wang Tianmiao

Hybrid undulatory kinematics of a robotic Mackerel ( Scomber scombrus ): Theoretical modeling and experimental investigation

Fishes that use undulatory locomotion occasionally change their inherent kinematics in terms of some natural behavior. This special locomotion pattern was vividly dubbed “hybrid kinematics” by biologists recently. In this paper, we employed a physical model with body shape of a Mackerel (Scomber scombrus), to use the three most typical undulatory kinematics: anguillform, carangiform and thunniform, to investigate the hydrodynamic performance of the so-called “hybrid kinematics” biological issue. Theoretical models of both kinematics and hydrodynamics of the physical model swimming were developed. Base on this model, the instantaneous force produced by fish undulatory body and flapping tail were calculated separately. We also quantitatively measured the hydrodynamic variables of the robotic model swimming with the three undulatory kinematics on an experimental apparatus. The results of both theoretical model and experiment showed that the robot with thunniform kinematics not only reaches a higher speed but also is more efficient during steady swimming mode. However, anguilliform kinematics won the speed race during the initial acceleration. Additionally, the digital particle image velocimetry (DPIV) results showed some difference of the wake flow generated by the robotic swimmer among the three undulatory kinematics. Our findings may possibly shed light on the motion control of a biomimetic robotic fish and provide certain evidence of why the “hybrid kinematics” exists within the typical undulatory locomotion patterns.

Attitude estimation for small helicopter using extended kalman filter

Accurate attitude estimation is vital for VTOL UAV (vertical take-off and landing unmanned aerial vehicle) systems, including small helicopter. The main focus of present research is to compile the principles of navigation and Kalman filtering, and their application and implementation towards UAV systems. This paper introduces a method based on EKF (Extended Kalman filter) to estimate small helicopter¿s attitude with little drift and low noise, which uses strapdown attitude as state update, and treats the attitude estimated by the bi-vector method as measurement update. By evaluating the EKF method with small helicopters real ground and flight sensor data, it shows good performance: stable, small error. So it is suitable for small helicopter attitude estimation application.

On the thrust performance of an ionic polymer-metal composite actuated robotic fish: Modeling and experimental investigation

In this paper, we theoretically predict and experimentally measure the thrust efficiency of a biomimetic robotic fish, which is propelled by an ionic polymer-metal composite (IPMC) actuator. A physics-based model that consists of IPMC dynamics and hydrodynamics was proposed, and simulation was conducted. In order to test the thrust performance of the robotic fish, a novel experimental apparatus was developed for hydrodynamic experiments. Under a servo towing system, the IPMC fish swam at a self-propelled speed where external force is averagely zero. Experimental results demonstrated that the theoretical model can well predict the thrust efficiency of the robotic fish. A maximum thrust efficiency of 2.3×10−3 at 1 Hz was recorded experimentally, the maximum thrust force was 0.0253 N, recorded at 1.2 Hz, while the maximum speed was 0.021 m/s, recorded at 1.5 Hz, and a peak power of 0.36 W was recorded at 2.6 Hz. Additionally, the optimal actuation frequency for the thrust efficiency was also recorded at the maximum self-propelled speed. The present method of examining the thrust efficiency may also be applied to the studies of other types of smart material actuated underwater robots.

Control of a self-assembly modular robot system over a wireless ZigBee network

This paper emphasizes the control problem of a self-assembly modular robot with wireless network load. A self-assembly modular robot driven by four micro DC motors is used as an example, which is named Sambot. Each motor is controlled by a separate microcontroller-ATMega8 for driving, and the ATMega8 also complete the encoder and sensor information collection, the left and right wheel drive motor are used for the control of linear and angular velocities of the robot modular, the other one motor is used for the docking-platform rotating between two robot modular, and the last one motor is used for the hook lock after successfully docking. A Zigbee network is implemented between each other modular robot uses TIs CC2430 chip with C51 microcontroller core for the realization of the underlying layer protocol stack and upper layer application, and a CAN Bus communication uses VP230 as CAN transceiver to connect with the CAN interface when they are docking locked. The wireless network controlled system and the communication network are simulated respectively with Matlab/Simulink and Truetime, and the mobile self-assembly robot control and communication system has been a very good real-time and lowjitter validation.

A Middleware Based Control Architecture for Modular Robot Systems

Robot technology has been widely used in industrial production and daily life. Unfortunately, due to the uncertainty of the environment and the diversity of functions, the complexity of the robot software is growing dramatically, in which a lot of repetitive works are done. To solve this problem, the modularization of robots and robot middleware has been proposed. From the perspective of modularization, interoperability, scalability and ease-of-use, this paper proposes a loose-coupled, service-oriented modular robot middleware (MoRoM) architecture. In this architecture, we define the standard interface to encapsulate services each functional component provides, realize system automatic configuration (plug and play), and provide a message notification mechanism as well as program frameworks. To verify the proposed middleware, we implement it using ACE-TAO, which is a real-time CORBA implementation, and test it in a modular robot system. The test shows that this architecture is suitable for improving reusability, interoperability and scalability of robot software.

Curriculum of Embedded System for Software Colleges

Embedded system has already been applied in numerous fields, such as the industry control system, information appliances, communication equipments, medical instruments, intelligence instruments and so on. As processors become more powerful, electronic products become more complex. Embedded system designers are able to include design features traditionally associated with personal computer (PC) including graphical user interfaces, TCP/IP networking, and database management. For introducing knowledge of embedded system to student of Master of Software Engineering (MSE), the Software College at Tsinghua University and Beihang University had an opportunity to add technical electives for embedded system, the author received approval to teach embedded system design course. This paper describes the development and implementation of that course toward the goal of teaching students the hardware considerations, software considerations, and design processes involved in embedded system design. In the end, the teaching effect of course for embedded system design is analyzed

An Attitude Estimate Approach using MEMS Sensors for Small UAVs

For the small UAVs (unmanned aerial vehicle) using MEMS sensors, this article puts forward a Kahnan Filter model to get attitude estimate without long term drift and showing relatively smaller error. Firstly, strapdown inertial attitude algorithm and bi-vector attitude algorithm are presented, which are widely used in small UAV autopilot systems now. However, there is a problem of long term drift with the former and heavy noise with the latter. Due to these shortcomings, accurate attitude control has not been achieved yet in small UAVs. In order to solve these problems, this paper gives out a Kalman filter model which fuses the two types of data into an optimal estimate of real attitude, and overcomes the shortages of both algorithms mentioned above. Simulation results show that this filter can be used to gain fairly good data for more accurate attitude control. Besides, compared with the filters already developed, this Kalman filter has a relatively low order and a loose architecture, which could be more easily adopted in an existed embedded computer system of small UAV.

Design and kinematics simulation for bionic crank-slider mechanism of jumping robot

Jumping locomotion is an ideal means of overcoming obstacles and traversing rough terrain. By taking inspirations from the locust, this paper presents the development and analysis of a novel crank-slider jumping mechanism. Firstly, the locust morphology is described and the posture of hindlimbs at take-off phase is analyzed. Base on that, a crank-slider mechanism is proposed to mimic the locust hindlimb. The mechanical analysis shows that the ground reaction force is similar to that of the locust during take-off stage, which reduce the possibility of premature lift-off and lays the foundation for developing the small jumping robot. Then, the designed robot employs elastic elements in the crank-slider mechanism, which is triggered by the segment-gear system. At last, its jumping performance is verified by kinematic modeling.

Reconfiguration Research on Modular Mobile Robot

The reconfigurable tracked mobile robot was developed for all-terrain recon. Based on the unique characters of reconfigurable robot, modular mechanical and electrical structure and control method were all discussed. The tracked robot was composed of four basic locomotion modules, three joint rods and two rotational articulations. By the different configurations of all the modules, the mechanical reconfiguration of the tracked vehicle was realized. The master-slave control system was adopted. Based on the ARM real time embedded control system, the reconfiguration was realized when the mechanical structure of the robot was reconfigured. The tracked robot was validated that it has the excellent adaptable ability in complex terrain and has easy and simple reconfigurable performance when it executed recon in the building interior simulating the real task

Modularization of Miniature Tracked Reconnaissance Robot

This paper analyzes the demanding background of the tracked reconnaissance robot, and its requisite performance is discussed in detail. According to the demand, the miniature tracked reconnaissance robot (MTRR) is developed in Beihang University. The purpose of this paper is to discuss the application of the MTRR system and its modularization design methods. The meanings of modularization are that: structure modularization, control system modularization and software modularization. The MTRR system of small size and light weight is characterized with reconfiguration and portable performance. Furthermore, the system is provided with waterproof and shock-resistant performance