Bumblebee RobotX Journal 2024

For RobotX 2024, Team Bumblebee introduces BBASV 4.0 and Jellyfish 2.0, the latest iterations of our Autonomous Surface Vessel (ASV) and Unmanned Aerial Vehicles (UAV). Major upgrades focus on improving reliability, modularity, and ease of maintenance. Key innovations include an enhanced telemetry system, optimized ball-shooter mechanisms, and inter-vehicle communications between the ASV and UAV. The team’s software stack, based on ROS 2, integrates sophisticated sensor fusion and navigation systems, enabling precise decision-making in dynamic maritime environments. This report highlights the design improvements, testing methodologies, and competition strategies for this year’s challenge.

Bumblebee RoboSub Journal 2023

For RoboSub 2023, Team Bumblebee’s strategy involves deploying an improved BBAUV 4.1 equipped with techniques and contingencies to effectively complete tasks at TRANSDEC. The mechanical work has upgraded BBAUV 4.1’s actuation capabilities with the introduction of the hydraulic vacuum gripper, compliant claw and improved torpedo hydrodynamics. Electrical work centered around implementing quality of life changes into the system, electrical workflow improvements and system reliability. Software development focused on improved accuracy and robustness of our perception pipeline, more convenient mission planning workflows and robustness of obstacle approach strategies.

Bumblebee RobotX Journal 2022

Bumbleebee Autonomous Surface Vessel 3.0 is the newest iteration of Team Bumblebee’s Autonomous Surface Vessel (ASV), and features major upgrades to the mechanical, electrical, and software subsystems. The integration of an Unmanned Aerial Vehicle in RobotX 2022 presented several challenges which are discussed in this paper, along with the various upgrades we have added to the ASV during our development process.

Bumblebee RoboSub Journal 2022

For RoboSub 2022, Team Bumblebee’s strategy involves deploying the BBAUV 4.0 to efficiently complete all tasks. The mechanical design of the BBAUV 4.0 optimises for space and weight, yielding vastly superior manoeuvrability while eliminating the weight penalty of the BBAUV 3.99. Electrical work centred on testing and integrating new sensors, as well as improving overall system reliability. Software development focused on an updated behaviour-tree mission planner, improvements to the controls system, and driver development for our new sensors. Physical testing was conducted at a steady pace, and was supplemented using hydrodynamic simulations to tune the AUV’s control systems.

Bumblebee RoboSub Journal 2021

For RoboSub 2021, Team Bumblebee’s strategy involves simultaneously deploying the BBAUV 3.99 and the BBAUV 4.0 to complete the tasks efficiently, using acoustic subsystems for inter-vehicular communication. The mechanical design of the BBAUV 4.0 is optimised for space and weight, offering vastly superior manoeuvrability while eliminating the weight penalty incurred by the BBAUV 3.99. Electrical work centred on improving ease-of-access to components and decoupling individual subsystems, reducing turnaround time when debugging. Software development focused on improving the vision pipeline and implementing a behaviour tree-based mission planner to allow for quicker iterations during testing and an easier understanding of the AUV’s behaviour. The team’s limited access to in-water testing was supplemented using hydrodynamic simulations to tune the AUV’s control systems, and custom sensor plugins were developed for our Gazebo simulator. Testing time was used to qualify our simulation results and improve our models.

Bumblebee RoboSub Journal 2020

BumbleBee Autonomous Underwater Vehicle (BBAUV) 4.0 is the product of a team of undergraduates from the National University of Singapore (NUS). BBAUV 4.0 is a ground-up redesign of the mechanical and electrical system, improving the performance and stability of the vehicle, as well as an increased focus on software control systems. This paper discusses the design of BBAUV 4.0 and the improvements over our previous vehicles.

Bumblebee RoboSub Journal 2019

BBAUV 3.5X is the product of a team of undergraduates from the National University of Singapore (NUS). BBAUV 3.5X aims to complete the tasks by playing to the team’s strengths: Acoustics Localization, Waypoint Navigation and Object Detection using Sensor Fusion. The vehicle was further improved over the previous version by redesigning the frame for better stability and protection, along with upgrades in the manipulator system, as well as increased focus on deep learning based computer vision. This paper discusses their integration and performances on the 3.5X.

Bumblebee RobotX Journal 2018

Bumblebee Autonomous Surface Vessel (BBASV) 2.0 is the product of a team of undergraduates and alumnus from the National University of Singapore (NUS). As part of the team’s 3-year masterplan, the BBASV will demonstrate its capabilities, along with the team’s flagship Bumblebee Autonomous Underwater Vehicle (BBAUV) 3.5, in a multi-platform system during the Maritime RobotX Challenge 2018. This paper discusses the development and integration work done by the team for the past 2 years.

Bumblebee RoboSub Journal 2018

Bumblebee 3.5 is the product of a team of undergraduates from the National University of Singapore (NUS). This vehicle is upgraded from 3.0 with considerations for RoboSub and Maritime RobotX. Specifically for RoboSub, Bumblebee 3.5 aims to complete the tasks by banking on our key strengths: Acoustics Localisation, Waypoint Navigation, and Object Detection using Sensor Fusion. Given the increased difficulty of the tasks, Bumblebee 3.5’s software stack is improved with new software frameworks, to include a revamped mission planner and integration of deep learning algorithms. This paper discusses their integration and performances on the 3.5.

Bumblebee RobotX Journal 2016

Bumblebee Autonomous Surface Vehicle 1.0 (BBASV 1.0) is an extension of the team’s work on autonomous maritime vehicles development, guided by experience gained from the development of the Bumblebee Autonomous Underwater Vehicle (BBAUV), deployed in RoboSub, another AUVSI Foundation Competition. This first iteration was developed over a short six-month design and implementation cycle, providing a baseline vehicle to be used for future development work. It features a baseline ASV system with obstacle avoidance, feature tracking, autonomous navigation and acoustic localisation capabilities.

Bumblebee RoboSub Journal 2016

Bumblebee Autonomous Underwater Vehicle (BBAUV) is the product of a team of undergraduates from National University of Singapore (NUS). This vehicle is designed for two competitions: AUVSI RoboSub Competition and the Singapore AUV Challenge. The Bumblebee 3.0 vehicle was fully modelled in CAD and fabricated with CNC machining, laser cutting and 3D printing. This year the vehicle features many improvements such as a more streamlined mechanical structure, reliable and fault tolerant electrical system as well as more accurate navigation. Sensor fusion between the imaging sonar and the fron-camera was implemented to fascilitate better object tracking and recognition. Bumblebee 3.0’s sensor suite includes a Doppler Velocity Log (DVL), an active imaging sonar, hydrophone array, inertial measurement units and two machine vision cameras. The software architecture is based on Linux, with Robot Operating System (ROS) framework as middleware.

Bumblebee RoboSub Journal 2015

BumbleBee Autonomous Underwater Vehicle (BBAUV) is the product of a team of undergraduates from National University of Singapore (NUS), title-sponsored by Hallin Marine. This vehicle is designed for two competitions: the 17th AUVSI RoboSub Competition and the Singapore AUV Challenge. The BumbleBee vehicle was fully modelled using the SolidWorks CAD package and fabricated by Cititech Engineering and NUS. BumbleBee presents a more modular and robust frame, and incorporates new advancements such as custom fabricated electrical boards and significant software changes for mission robustness. BumbleBee’s sensor suite includes an Explorer DVL, an imaging sonar, a hydrophone array, an IMU, two colour cameras, a depth sensor and an internal pressure sensor. Its software architecture is built upon ROS Hydro and more complex vision algorithms have been implemented using OpenCV Python.