The purpose of the proposed work was to develop an autonomous ship model. In this manner an RC model was fitted with the necessary equipment and tested in the towing tank of the National Technical University of Athens (NTUA). The goal was to develop and evaluate our system before installing similar equipment on larger boats for testing at sea. The work started as a continuation of the PhD thesis of the third author. The hull of the model was based on an RC high speed model. Two DC brushed motors paired with shafts and propellers were used for the propulsion. For the power supply of the model, two battery packs were assembled for the motors and the controller respectively. Step-down circuits were used to match the sensors’ requirements. A servo motor was utilized for the rudder control. Six proximity sensors and a LiDAR were fitted to the model’s deck in order to capture the surrounding environment. Moreover, a 6-Dof IMU that combines data from three different sensors (magnetometer, gyroscope, and accelerometers) provided the model’s motions. A GPS module was fitted in the model for future use in an open environment. RPM meters were fitted to the propeller axes to control the model’s velocity. The position of the model in the towing tank was acquired by an optical real time system. For the main control of the model, a controller devised and manufactured by the third author of this paper was used for the RC communications, rpm meter and thrust control. The rest of the processes were executed by a Raspberry Pi 5 computer. A Raspberry Pi Pico microcontroller was adopted for the motion control of the rudder. Different approaches were developed and tested for the optimum vessel control. Python, C and C++ programming languages were used to develop the software to control the model. Several modules were implemented to the main script depending on the nature of the experiment. In any case, the system acquires the data from the sensor and calculates the optimum commands for the model’s navigation. Finally, all the data are stored to a portable USB device for post processing.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Development and Evaluation of an Autonomous Ship Model in the Towing Tank of NTUA

  • S. P. Papadopoulos Kouklakis,
  • D. E. Liarokapis,
  • E. C. Loghis,
  • G. P. Trachanas,
  • G. J. Grigoropoulos

摘要

The purpose of the proposed work was to develop an autonomous ship model. In this manner an RC model was fitted with the necessary equipment and tested in the towing tank of the National Technical University of Athens (NTUA). The goal was to develop and evaluate our system before installing similar equipment on larger boats for testing at sea. The work started as a continuation of the PhD thesis of the third author. The hull of the model was based on an RC high speed model. Two DC brushed motors paired with shafts and propellers were used for the propulsion. For the power supply of the model, two battery packs were assembled for the motors and the controller respectively. Step-down circuits were used to match the sensors’ requirements. A servo motor was utilized for the rudder control. Six proximity sensors and a LiDAR were fitted to the model’s deck in order to capture the surrounding environment. Moreover, a 6-Dof IMU that combines data from three different sensors (magnetometer, gyroscope, and accelerometers) provided the model’s motions. A GPS module was fitted in the model for future use in an open environment. RPM meters were fitted to the propeller axes to control the model’s velocity. The position of the model in the towing tank was acquired by an optical real time system. For the main control of the model, a controller devised and manufactured by the third author of this paper was used for the RC communications, rpm meter and thrust control. The rest of the processes were executed by a Raspberry Pi 5 computer. A Raspberry Pi Pico microcontroller was adopted for the motion control of the rudder. Different approaches were developed and tested for the optimum vessel control. Python, C and C++ programming languages were used to develop the software to control the model. Several modules were implemented to the main script depending on the nature of the experiment. In any case, the system acquires the data from the sensor and calculates the optimum commands for the model’s navigation. Finally, all the data are stored to a portable USB device for post processing.