Adaptive Neuro-Fuzzy Active Spring Control of a Vibration Isolator for Towed Capacitive Resistivity Imaging Single Trailer

Underground imaging land surveying is oftentimes associated with vibrations and ground excitation due to the presence of rough and uneven road surfaces. In addition, the requirement of the system to maintain direct contact with the ground or with 5 mm maximum elevation during resistivity data gathering remains a challenge. With that, a vibration isolator should be considered especially with a semi-active control system that enables the real-time operation of mitigating vibrations and a function that enables the manipulation of the electrode holder to maintain ground contact. Thus, this study aims to focus on the development of an active spring control system by utilizing and comparing fuzzy logic and ANFIS to be employed in a semi-active vibration isolator for a towed capacitive resistivity imaging application. By installing a vibration isolator between the dipole antenna and trailer frame, it can control the displacement of the horizontal spring resulting in the force that will enable the lifting and lowering of antenna dipoles. The crips inputs included in this paper are electrode holder resultant force and displacement as well as the top trailer frame displacement while spring displacement for the crisp output. There is a total of twelve generated fuzzy rules that were established for fuzzy logic. On the other hand, an adaptive neuro-fuzzy inference system (ANFIS) was also employed for performance comparison with fuzzy logic that utilized gaussian membership functions and generated twenty-seven rules. Based on the comparison of results and model evaluation metrics, ANFIS has a more stable, accurate, and controlled performance that could be integrated with a servo-motor controller system for the intended semi-active vibration isolator.