other controller design

Model identification and loop-shaping based controller design for VTOL UAV

In the past, we used the traditional PID controller in the attitude loop, the controller is stable but the performance is not very good. So we tried to design the controller in a more systematic approach. The simulation results show that the angular rate controllers designed by model identification and loop-shaping method have better performance than the controllers designed by PID method, mainly in the following areas: (1) better time domain response: less overshoot, shorter rise time, etc.; (2) More intuitive to design the controllers according to the Bode diagram, achieving good gain margin, phase margin and bandwidth; (3) This method allows us to analyze the signal response in different frequency bands, and design a more robust controller, which can suppress noise interference effectively.

Students

PhD students

Haowei GU 顾昊伟
February 2018- Present
Bachelor in Xi’an Jiaotong University
Research focus: Aircraft design and optimization

Hao XU 徐浩
February 2018 – Present
B.Eng. in University of Science and Technology of China

Jinni ZHOU 周晋妮
September 2014 – Present
Bachelor and Master in Central South University
Research focus: VTOL control and motion planning

Mphil students

Xiaoyu CAI 蔡潇宇
September 2017- Present
Bachelor in Xi’an Jiaotong University
Research focus: Aircraft design and control

Alumni

Ximing LYU 吕熙敏
September 2014- August 2018
Bachelor in Harbin Institude of Technology
Research focus: VTOL design and control
 
 
 
 
 
 
 
Ya WANG 王亚
September 2014- May 2017
Bachelor in Xi’an Jiaotong University
Research focus: Aircraft design and mechanical analysis
 
 
 
 
 
 
 
 

Environment monitoring

Environment issues play an important role in judging the potential of a region. Emission standard is made to constrain the pollution and officers equipped with monitor facility are sent to spot check the factories one by one. Our platforms can be used to fly among different factories and do the spot check mission which will be quick and quiet. So we can increase the success rate to find out which factory is actually doing the illegal pollution.

Disaster relief & rescue

Most of the disaster sites are complex and the affected area is quite large. It is usually hard for fixed wing with searching equipment to find suitable take-off and landing field. It is also hard for rotary-wing to cover the affected area within reasonable time and sorties. In this scenario, our VTOL platform with high power efficiency can fulfill the task efficiently without considering take-off and landing condition. More lives may be saved because of the usage of our platform.

 

Parcel delivery

Every year, tourist resorts in Hong Kong attract millions of visitors all around the world. A high percentage of the sights are islands, country parks and walking trials. The common features of these places are that most of them are far away from hospital, ground transportation is inaccessible. In reality, helicopters are used to transport the patient to hospital. This is a safe and quick solution, except the extremely high expense. In many cases, what the patients need is emergency medicine and some first-aid goods. If our platform can be merged to the emergency service system and used to transport these staffs more people can benefit from the low expense emergency medical service.

Design & optimization

Design

VTOL UAVs design means the comprehensive system development of such VTOL UAVs from all aspects, including the aircraft design and implementation, onboard devices integration, ground station support, and long-distance communication. We proceed with aerodynamic analysis, mechanical design, and manufacture.
Up to now, we designed 2 versions of combine VTOL UAV and 2 versions of tail sitter VTOL UAV. And our present work is that design the third version of combine VTOL and the third version of tail sitter VTOL UAV.

Optimization

To reach the maximum range or endurance, we optimized VTOL system design, include aircraft layout, aerodynamic parameters specification and propulsion system selection. We write a MATLAB program to optimize this problem where some experiential or theoretical formulas simplified the whole VTOL system. The flight range or endurance is taken as objective function.
This optimization program can be used for any kind of VTOL aircraft design by specified the number of propeller, maximum take-off weight and battery parameters.

 

unified control method

A unified control method for quadrotor tail-sitter UAVs in all flight modes: hover, transition, and level flight.

On the current quadrotor tail-sitter UAV, we used 2 different control schemes, quadrotor mode and fixed-wing mode. Although this platform can finish an autonomous flight based on some waypoints, but it has the altitude drop when performing transition flight. This altitude drop problem is very common among most of the VTOL platform.
Therefore, a new control scheme is proposed, it treats the hovering and forward flight mode uniformly, and enabling continuous transition between these two modes, depending on the commanded velocity. The simulation results are quite good, and this part of work has been accepted by IROS 2017, the title is “A Unified Control Method for Quadrotor Tail-sitter UAVs in All Flight Modes: Hover, Transition, and Level Flight”. The experiments and simulation results can be seen in the following video.
Video
 

Control

                                                                                   Visualization Transition

VTOL control focus on the system identification, model based simulation and controller design.  To achieve these, full scale vehicle wind tunnel tests are conducted to get the aerodynamic of our vehicle; motor wind tunnel tests are conducted to get the performance of our rotors according to advance ratio and a hierarchical controller structure is built to do both manual and autonomous flight.

Control targets include:

  • quad rotor tail-sitter;
  • dual rotor tail-sitter;
  • hybrid system vehicle.

Current research interests including: online and off-line system identification, wind estimation, adaptive control, disturbance observer design.
 

Border patrol

China has the longest land border around the world with more than 22,000 kilometers in total. Many soldiers (official number is unavailable) are sent to guard and protect the border. Considering the complex landing conditions and long distance between sentry posts, our VTOL UAV platform is suitable for the patrol mission.

It would be safer and more efficient if the VTOL UAV platform is widely used in the border patrol mission. Similarly, there is a 30-km border Hong Kong and the main land. The VTOL UAV platform can be used to prevent smuggling, drug trafficking, illegal immigration, etc.

Web description

A vertical takeoff and landing (VTOL) unmanned aerial vehicle (UAV) employs a wing to produce its lift like a fixed-wing air- plane, and at the same time, is capable of vertical takeoff, landing and hovering, like a multi-copter. In this way, a VTOL UAV obtains both flight efficiency and agility.

Investigators

Prof. Shaojie SHEN


Assistant Professor, Dept. of Electronic & Computer Engineering, HKUST
Director, HKUST-DJI Joint Innovation Laboratory

Professor Shaojie Shen received his B.Eng. degree in Electronic Engineering from the Hong Kong University of Science and Technology in 2009. He received his M.S. in Robotics and Ph.D. in Electrical and Systems Engineering in 2011 and 2014, respectively, all from the University of Pennsylvania. He joined the Department of Electronic and Computer Engineering at the Hong Kong University of Science and Technology in September 2014 as an Assistant Professor. His research interests are in the areas of robotics and unmanned aerial vehicles, with focus on state estimation, sensor fusion, computer vision, localization and mapping, and autonomous navigation in complex environments. He and his research team won the best paper finalist in ICRA2011, best paper award in SSRR2015 and SSRR2016, and first price in IARC2015. He was the program co-chairs of SSRR2015 and SAC2017-IRMAS.

Prof. Chihyung WEN


Associate Head and Professor, Dept. of Mechanical Engineering, HKPolyU
Associate Fellow of AIAA; Fellow of HKIE

Professor Wen received his Bachelor of Science degree from the Department of Mechanical Engineering at the National Taiwan University in 1986 and Master of Science and PhD from the Department of Aeronautics at the California Institute of Technology (Caltech), U.S.A. in 1989 and 1994 respectively. He had been the Chairman of the Department of Mechanical and Vehicle Engineering from August 1997 to July 2000, and the Provost from August 2004 to July 2006 in the Da-Yeh University, Taiwan. In August 2006, Professor Wen joined the Department of Aeronautics and Astronautics of the National Cheng Kung University (NCKU), Taiwan, before joining the Department of Mechanical Engineering, The Hong Kong Polytechnic University in 2012 as professor. His current research interests are in the areas of (1) Aerodynamic applications of plasma actuators in delta-winged UAVs and MAVs; (2) Hypersonic aerodynamics and scramjet engine design; (3) Fuel cell applications in the electric power system of a micro spacecraft; and (4) Flow instabilities of magnetic fluids and their applications in micro-mixers.

Prof. Fu ZHANG


Research Assistant Professor, Dep. of Electronic & Computer Engineering, HKUST

Professor Fu Zhang received the B.E. (with honor) from the Department of Automation, University of Science and Technology of China, Hefei, China,in 2011. He received his Ph.D. in Mechanical Engineering, the University of California, in 2015, with research focusing on the dynamics modeling and control of ultra-high precision machining and high performance micro-scale rate integrating gyros. Fu Zhang is a research assistant professor with the Department of Electronic and Computer Engineering at the Hong Kong University of Science and Technology since August 2016. His current research interests include the dynamics, control, planning and navigation of aerial robots, deep reinforcement learning and optimization.