Revamp Software Quality for Android Robots through Advanced Development and Deployment Methodologies
Robotics software quality is crucial to control robot hardware while interacting with the physical world to accomplish a set of complex tasks. Software is evolving to tackle sensor errors, robot software/hardware failures, human errors, and uncertainties from the external environment. This paper describes challenges to revamp the software quality for an Android platform robot based on the experiences of the First Tech Challenge (FTC) robotics team, Revamped Robotics. The FTC competition requires a three-month period between the competition challenge release and the first competition. The FTC season is roughly eight months consisting of multiple rounds of robot performance improvements, as well as FTC competitions. To revamp the quality of work in the FTC season, the team employed SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis methodology to enable smooth coordination between the hardware and software aspects of the robot, ensuring that they operate efficiently and reliably. The team utilized agile development processes, the cloud-based GitHub software management and CI/CD methods enabling frequent software improvement cycles. Furthermore, deep learning TensorFlow based neural networks were implemented to enable accurate autonomous object detections to be processed by the software, allowing for consistent robot movements.
Authors are high school students from the Revamped Robotics team. The team is a six-year veteran FIRST Tech Challenge (FTC) team from Portland, Oregon. Through their six-year journey as a team, they have participated at the state and world level competitions. They advanced to the world championships three times and received recognitions for their competitive robot, programming and contribution to the communities. The team developed a fast and streamlined design and building process while ensuring software quality. In addition, the team also focuses on outreach activities in educating young students in communities about Science, Technology, Engineering and Mathematics (STEM) through Tech Talks, community open houses, and mentoring younger robotics teams.