Project Director

Ofoli, Abdul R.

Department Examiner

Loveless, Thomas D.; Ahmed, Raga


Dept. of Electrical Engineering


University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)


Fuzzy logic seeks to express human modes of reasoning and decision making in a mathematical form. This is evident in its terminology such as “linguistic variables” defined over a “universe of discourse”. By taking human expressions such as “very high” or “pretty cold” and defining them in a mathematical context, expert operator knowledge can be transferred from verbal descriptions into automated control algorithms regardless of the operator’s familiarity with control systems. Because fuzzy logic is designed to be easily comparable with human thought, it makes an excellent first exposure to control systems concepts to high school and undergraduate students. Additionally, one of the barriers preventing widespread industry use of fuzzy controls is that the emerging workforce is not familiar enough with fuzzy controls to successfully operate a fuzzy system. This work will demonstrate the suitability of fuzzy controls for education at the undergraduate level through the development of BasketBallBot. BasketBallBot uses the educational platform distributed to high schools throughout the country through the FIRST (For Inspiration and Recognition of Science and Technology) Robotics Competition (FRC). Inexpensive sensors are added to the robot and interfaced using the easily accessible Arduino platform. The affordability of the sensors and prevalence of the computing and hardware platforms insure that this work could be recreated at other undergraduate institutions and even high schools. This paper will thoroughly describe the sensor integration process. It also describes a heuristic technique for developing fuzzy logic controllers and inference systems that does not require a high level of mathematics to use. This technique is employed to design several controllers and a fuzzy inference system. These controllers’ performance is investigated through simulation and experiment. Finally, the fuzzy inference system is developed that prescribes the desired ball launch speed given the distance to the hoop.


The authors would like to acknowledge Brady, McNabb, Broadstone, and Borden for their work on the camera aiming software for BasketBallBot. Thanks to Dr. Ahmed and Dr. Loveless for sitting on the review committee and providing feedback on drafts. Huge thanks to Dr. Ofoli for guiding this research so kindly and informatively; without your scope on the project I would have wandered terribly. Thanks to Brian MacCleery from National Instruments for providing us with the NI software license. Many thanks to the UTC robotics team for lending their tools and when it was needed most.


B. S.; An honors thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Bachelor of Science.




Fuzzy logic; Robotics; Robots; Detectors; Fuzzy systems


robot; artificial intelligence; controls; fuzzy controls; STEM education


Electrical and Computer Engineering

Document Type



iii, 64 leaves







BasketBallBotMontage.mp4 (10479 kB)
Video of BasketBallBot shooting basketball hoops. At first it is untrained and misses 50% of shots. After training it makes 100% of shots until the robot is moved far away from the hoop at the end.