List of papers

• A New Turbine-Sail Coupled Propulsive System for Autonomous Sailboats,Jiayi Qiu, Jiafan Hou, Chongfeng Liu, Hengli Liu, Xiongwei Lin, Zhenglong Sun, Ning Ding, Tin Lun Lam and Huihuan Qian
• Collaborative Object Transportation by Multiple Robots with Onboard Object Localization Algorithm, Zhixian Hu, Zhixiang Zhao（coordinate first author）, Lianxin Zhang, Hengli Liu, Ning Ding, Zhenglong Sun, Tin Lun Lam, Huihuan Qian
• An Adaptive Position Keeping Algorithm for Autonomous Sailboats, Zeyuan Feng, Jiayi Qiu, Hengli Liu, Qinbo Sun, Ning Ding, Zhenglong Sun, Tin Lun Lam and Huihuan Qian 
• Wing Sail Land-yacht Modeling and System Verification, Yibing Dong, Xiao Ding, Zhijun Li（coordinate first author）, Lianxin Zhang, Hengli Liu, Ning Ding, Zhenglong Sun, Huihuan Qian 
• Obstacle Avoidance for Autonomous Sailboats via Reinforcement Learning with Coarse-to-fine Strategy, Ziyuan Cheng, Weimin Qi, Qinbo Sun, Hengli Liu, Ning Ding, Zhenglong Sun, Tin Lun Lam and Huihuan Qian

Introduction of ROBIO

IEEE International Conference on Robotics and Robotics (ROBIO) has been successfully held for 15 years. It is a well-known international conference on robotics in Asia Pacific region. The ROBIO 2019 has be held in Dali, Yunnan, China in December.

About the papers

Under the guidance of Assistant Professor Qian Huihuan, Qiu Jiayi and Hou Jiafan, grade 16 electronic and information engineering majors, and Liu Chongfeng, grade 19 phD students, completed "a new turbine sail coupling propulsion system for autonomous sailing ships". In the ocean voyage, unmanned sailboats need to solve the energy problem and make better use of wind energy to obtain the required propulsion. In this paper, a new system with vertical generators on both sides of the sail and controllable speed of the generator is proposed to explore the influence of the system on the maximum propulsion of the sailboat. According to a large number of experiments completed in the wind tunnel laboratory, it is found that the system can increase the maximum propulsion force of the sailboat when sailing downwind. It is also confirmed that the maximum propulsion force can be further adjusted by adjusting the speed of generators on both sides.

Under the guidance of Sun Zhenglong, a research assistant professor, and Zhang Lianxin, a phD student, Hu Zhixian and Zhao Zhixiang, majoring in electronic information engineering of level 16, completed the collaborative transportation of multiple robots based on airborne lidar positioning algorithm. In this paper, a new airborne lidar positioning algorithm is proposed, and a common cooperative handling control strategy is designed. The effective combination of the algorithm and the strategy can realize the object trajectory tracking in the process of being transported. In the previous studies of collaborative transportation, the positioning of the object to be carried depends on additional positioning device or fixed device. The positioning algorithm proposed in this paper, without the help of additional devices, can also accurately determine the position of the object to be carried. When the position of the object is known, the cooperative handling control strategy can effectively control the transportation process. The experimental results in real scene verify the effectiveness of the localization algorithm, and further experiments prove the feasibility of the collaborative transportation strategy.

Under the guidance of postdoctoral Liu Hengli and phD student Sun Qinbo, Feng Zeyuan, a 17 level electronic information engineering major, completed the "adaptive sailing position keeping algorithm". Ship position keeping is very important in practice, such as positioning unmanned underwater vehicle, exchanging cargo, virtual anchoring and so on. Due to the characteristics of sailing wind driving, it is quite challenging to maintain its position. The current algorithm usually ignores the environmental disturbance, especially the wind disturbance when the sailboat is loose, so the effect is not good in the actual scene. In this paper, a more robust algorithm and underlying controller are proposed to keep the position of the sailboat under the condition of wind speed and wind direction changing. The control effect is verified on the simulation and experimental platform.

Under the guidance of phD student Zhang Lianxin, the paper "modeling and system verification of wing sail vehicle" jointly written by Dong Yibing, Ding Xiao and Li Zhijun, grade 17, majoring in electronic information engineering, puts forward a practical structural design of wing sail vehicle, models and verifies it. In the field of land sail vehicle, the commonly used soft sail vehicle has poor stability, and has greater limitations in the direction of travel. However, the research on the wing sail vehicle is often limited to the ability to run along the wind.This paper presents a sail car system based on hard wing sail. The four-wheel wing sail vehicle has good stability and controllability, and has a large range of driving directions. At the same time, the modeling and simulation experiments on the wing sail as well as the experiments in the real wind field verify the good driving ability of the wing sail vehicle in the multi angle upwind condition.

Under the guidance of phD students Qi Weimin and Sun Qinbo, Cheng Ziyuan, a 16 level computer science and technology major, proposed an autonomous obstacle avoidance method for sailing ships from coarse-grained to fine-grained based on reinforcement learning in his paper "unmanned sailboat obstacle avoidance based on reinforcement learning". As a very important input of the ship's position information, more accurate positioning can effectively improve the collision avoidance efficiency. Specifically, the coarse-grained boat can be roughly positioned, and then the boat can jump to the fine-grained one to obtain a more accurate position. The coarse-grained stage is used to control the normal sailing of the sailboat when it is far away from the obstacles, while the fine-grained stage enables the sailboat to avoid the obstacles accurately when it is close to the obstacles. We not only verify the feasibility and effectiveness of the algorithm through simulation, but also test it in the real experiment, and achieve good avoidance effect. This paper has obtained the best paper final list award.