ROBOBOAT 2023

RoboBoat is an international student competition to design autonomous robotic boats. The robotic boat navigates through the challenge course with impressive maritime maneuverability. This year is the second year for Bengawan UV Roboboat team to compete in the Roboboat competition.

ABOUT INTERNATIONAL ROBOBOAT COMPETITION

Our

MANDAKINI catra

Mandakini Catra is the third ship from Bengawan UV Roboboat Team that made for the 2023 International Roboboat Competition (IRC). Mandakini Catra is a development from previous two ships with modular concepts. De velopment from this ship’s is on program algorithms, object detection readings, fully programme d navigation system, and propulsion systems. Mandakini catra is developed with many new things because We believe big change starts with trying new things.

Technical design report

TEAM INTRODUCTION video

DEVELOPMENT TIMELINE

Mandakini Catra is designed using Maxsurf and Fusion 360 software. Mandakini Catra uses a symmetrical Catamaran hull type because it has good stability. The hull on the Mandakini Catra is produced with more fiberglass and gel coat layers compared to the Mandakini EVO so it is more rigid and does not require additional structure on the Mandakini Catra body.

hull design &
manufacturing

Electrical system
diagram

Mandakini Catra came up with a new system and performance, using a fully-programmed system improving last year system that still uses a mechanical system using relays. A detection system that uses Object Detection will certainly improve detection performance and a navigation system that can be controlled using PID which presents a more accurate Mandakini Catra. Supported by an emergency protection system from team development, Mandakini Catra is ready to complete the mission perfectly!

SIMULATION

Hull analysis

Mandakini Catra hull testing using ANSYS and Maxsurf software. The team conducted stability testing and Response Amplitude Operator (RAO) aimed at analyzing the ship's ability to return to it s equilibrium point as well as the ship's motion response to dynamic water waves.

Stability analysis shows that Mandakini Catra has an advantage in terms of stability over Mandakini EVO. With a turning degree of 30° and a maximum GZ value of 0.1067 m, Mandakini Catra is more stable than Mandakini EVO which only has a turning degree of 20° and a maximum GZ value of 0.1747 m under the same simulation conditions.

Frame analysis

The team conducted a static stress test to determine the deformation and value of the safety factor that occurred in the frame. Testing using Fusion 360 software. Mandakini Catra frame is designed to meet the value of safety factor greater than 4.0 so that safety is guaranteed and able to withstand a load of 11.9 kg

From the data of static stress simulation results, it can be concluded that the T-slotted profile aluminum frame meets the safety factor value with a minimum value of 4.77. The frame on the Mandakini Catra only undergoes very small deformations so that it is safe to withstand the weight of the ship.

Experimental

propulsion system

In propulsion testing, the team conducted thrust and maneuverability tests. Thruster testing is carried out directly in the water by hooking the frame on the ship's prototype with digital scales. Then, Mandakini Catra was given a maximum speed on the bike for five tries using the T200 Blue Robotics Thruster. Based on these tests, a maximum thrust value on the ship was obtained of 10.86 kg.

Water blaster

Water blaster testing is carried out directly on land. The nozzle of the water blaster is placed about 1 meter from the target. The test results showed that the water blaster can shoot water up to a distance of 3.2 meters at an angle of 38° from the water surface. After that, water is shot at the target hole. The average time it takes to fill a water blast tank is about 20.93 seconds.

object detection

Mandakini Catra uses an object detection algorithm with Tensorflow Lite. In the Tensorflow Lite framework, there are five architectural models that can be used to train custom models. We tested the sample data and the number of step trains of the five architectural models. The data obtained from the test is in the form of average Frames Per Second (FPS) and mean Average Precision (mAP)

Based on the test results, the architectural model that gets the highest average FPS result is EffidentLite0 at 7.6 FPS and mAP at 25.69 %. While the architectural model that gets the lowest average FPS result is EffidentDet-Lite4 at 0.8 FPS and mAP at 41.96 %. Therefore, we decided to use the EffidentLite0 architecture model so that the Raspberry Pi 4 device can run optimal object detection when performing missions.

DEVELOPMENT progress

PLANNING STRATEGY AND RESEARCH
We conducted an evaluation the previous year and then conducted discussions and conducted research to prepare for the IRC 2023 strategy. Based on the results of discussion and research, we decided to use a symmetrical hull catamaran with the concept of modularity to make it easier for us to carry the ship when traveling.

DESIGNING HULL AND FRAME
Furthermore, the ship design process is carried out. We do stability and RAO simulations on the hull and static stress simulations on the Frame to get the best hull and frame of the ship.

HULL MANUFACTURING
Hull vessels are made from fiberglass and gealcoat. First we make the hull out of plywood. Then we coat the hull with gealcoat to make a hull mold. Furthermore, the mold is coated with gealcoat and fiberglass to make the ship's hull. Next is painting using spray paint. After that, the installation of the frame on the hull

ELECTRICAL ASSEMBLY
We installed electronic components and propulsion systems on the ship. We arrange the placement of components appropriately so that electronic components are safe and can work optimally

SHIP TESTING
After the hull was completed, we conducted a buoyancy test to find out the ship's load and performed a maneuver test to determine the ship's response when turning.

TRIAL AUTONOMY CHALLENGE TASKS
After the ship and ship trials are ready for use, we will conduct ship experiments in accordance with the mission on Roboboat 2023.