Team Name
Bellstrike Squad
Timeline
Summer 2023 – Fall 2023
Students
- Jonathan Anderson
- Trenton Dragon
- Antonio Buentello
- Sho Reagan
- Chandni Patel
Sponsor
Bell Flight
Abstract
The Bell 2023 Advanced Vertical Robotics (AVR)Competition project was focused on the development and enhancement of the AVR Control System Fields for the recon, rescue, and firefighting challenge.
This project aimed to address various challenges and to improve the capabilities of the AVR Drones’ autonomous waypoint following.
By collaborating with Bell Engineers, the project aimed to create a robust system that could accurately and autonomously navigate waypoints using advanced computer vision and machine learning techniques.
Background
The competition aims to showcase how robotics and engineering can assist first responders in critical, life-threatening situations. The main goal is to develop a system that improves the autonomous waypoint-following capability of the AVR Drone. This will allow the drones to navigate more accurately and efficiently in GPS-denied environments, which will improve their performance in the AVR Competition and will in theory translate to performance improvements in real-world applications.
Such a system could also be used for other applications such as training, scouting, and research. Our goal was to improve the capabilities of the drone and its subsystems and to assist with the AVR competition to cultivate and inspire the next generation of aviation and computer engineers.
Project Requirements
WAYPOINT FOLLOWING IMPROVEMENT: The drone must demonstrate improved autonomous waypoint-following capabilities during the competition. The drone should be able to accurately navigate a predefined path of waypoints without human intervention. The system will use computer vision and machine learning techniques to identify waypoints and control the drone’s movement. Camera, thermal sensor, and autonomous code will be utilized to enhance waypoint recognition when GPS signals are unreliable indoors.
FUNNEL DESIGN IMPROVEMENT:The funnel design on the open-roof of the competition fields needs to be improved to prevent hollow balls representing water from rebounding upon reaching the aperture. The team plans to replace the current material of the open roof with a more suitable alternative, ensuring the smooth flow of the competition.
STAGE CONSTRUCTION AND FEEDBACK: The stage that was built for the competition needs improvements before the date of the competition and the customer will also need assistance with the construction of the new stages. We will be assisting by giving feedback to the current design and will help build the new stages with the suggested improvements.
WAYPOINT FOLLOWING SPEED AND ACCURACY: The AVR Drone must demonstrate a specific minimum speed and accuracy for autonomous waypoint following during the competition. The drone should be able to navigate between waypoints at a speed of at least 2 meters per second (m/s) while maintaining accuracy and obstacle avoidance capabilities.
DRONE HANDLING: During the operation or flight of the drone, all participants and observers must maintain a safe distance of at least 15 meters away from the drone at all times. The drone will be making quick and unpredictable movements as it navigates between waypoints, which could pose a potential safety risk to individuals in close proximity. Maintaining a safe distance is essential to prevent any accidents or injuries during the drone’s operation.
COMPETITION COURSE NET SAFETY: The competition course will be surrounded by a safety net to contain and protect the drones during the competition. The safety net will serve as a physical barrier to prevent the drones from flying outside the designated competition area and reduce the risk of collisions with spectators, participants, or other objects outside the course. The net will be securely fastened and properly tensioned to ensure it remains stable throughout the competition.
DRONE BATTERY CHARGING SAFETY: During the charging of drone batteries, it is essential that a responsible team member or designated individual be present at all times. Charging lithium polymer (LiPo) batteries used in drones carries inherent safety risks, such as the potential for overheating, fire, or explosion. Having a responsible person present during the charging process allows for immediate response in case of any battery-related incidents or emergencies. The charging process must be carried out in a well-ventilated area away from flammable materials.
PRODUCT MAINTENANCE AND TROUBLESHOOTING: The product maintenance and troubleshooting plan for the autonomous waypoint following, funnel design, and waterball pick up device will be designed to ensure consistent and reliable performance. This includes routine inspections, software updates, and hardware checks to identify and rectify any issues that may arise during its field use. A detailed troubleshooting procedure will be developed to promptly diagnose and resolve technical challenges, ensuring minimal downtime and maximizing the product’s uptime.
WAYPOINT FOLLOWING SOFTWARE PACKAGING: The software required for autonomous waypoint following of the AVR Drone is being shared to all teams involved. It is highly modular code that can be summed up as plug and play, since end-users are high school students.
PROVIDE SUPPORT FOR THE 2023 AVR Q&A FORUM: The Bellstrike Squad will actively participate in the AVR Q&A Forum during the 2024 competition season. The team will respond to inquiries and provide support to other teams by sharing insights, clarifying rules, and assisting in troubleshooting issues related to the competition.
System Overview
The way-point following system will be implemented using a combination of computer vision and machine learning techniques. The computer vision system will be used to identify the way-points, and the machine learning system will be used to control the drone’s movement to the way points. A sensor like LiDar will be implemented to be used as input for the AI. The system will be designed to be robust to noise and occlusion, unlike indoor GPS, and it will be able to handle a variety of different way-point configurations.
The funnel design will be changed to a more conical shape with a smaller aperture. This will help to reduce the amount of ball rebound and ensure that the balls flow smoothly through the funnel. The material of the open roof will also be changed to a more flexible material that will absorb the impact of the balls. The material of the open roof will be replaced with a more suitable alternative that mitigates ball rebound.
Major system components: A computer vision system to identify the way-points, machine-learning system to control the drone’s movement to the way-points, drone to carry the hollow balls, specialized device to pick up the hollow balls, funnel to guide the hollow balls into the designated area The external system interface will allow the system to be integrated with other systems, such as a scoring system or a live video feed. This will allow the system to be used in a variety of different applications, such as training, competition, and research.
Results
Results text and demo videos go here
Future Work
Future work text goes here
Project Files
System Requirements Specification
Architectural Design Specification (link)
Detailed Design Specification (link)
References
- Bell Textron & Robotics Education and Competition Foundation. (2022). Bell AVR competition logo. https://roboticseducation.org/bell-advanced-vertical-robotics/
- Robotics Education and Competition Foundation. (2022, December 3). BELL AVR 2022 championship event. Flickr. https://www.flickr.com/photos/recf/albums/72177720305473587/with/52645825291/
- Robotics Education and Competition Foundation. (2022b, December 3). BELL AVR 2022 championship event. Flickr. https://www.flickr.com/photos/recf/52645312272/in/album-72177720305473587/
