Multidimensional Force Sensor Dynamic Compensation Based on Multistrategy Improved Sparrow Search Algorithm

Multidimensional force sensors are extensively used in industrial milling machining for measuring cutting force and in intelligent robot applications for measuring joint force. These sensors offer several advantages, including improved static characteristics, well-established static calibration techniques, and temperature compensation technology. However, with the increasing demand for measuring dynamic forces in various applications, force sensors need to possess enhanced dynamic characteristics. Unfortunately, strain force sensors typically exhibit low intrinsic frequency and damping ratio, resulting in slower dynamic response of the sensor. To address this issue and enhance the dynamic performance of multidimensional force sensors, this study proposes a dynamic compensation method based on an improved sparrow search algorithm (ISSA). Chebyshev chaotic mapping was implemented to increase randomness and ergodicity in the initial population. An adaptive weight factor was incorporated to improve the position update formula of finders and the ratio of vigilantes. These changes enhanced the algorithm’s ability to conduct early global searches and late local depth mining. Subsequently,
${t}$
-distribution changes and Chebyshev chaotic perturbations were introduced to expand the local search capability. The enhanced sparrow search algorithm (SSA) improves the optimization capabilities of the original algorithm. Using dynamic calibration experimental data from 3-D force sensors, the algorithm’s effectiveness was verified. The results indicate that the method successfully reduces the overshooting amount in each channel of the sensors and shortens the regulation time. Consequently, the dynamic performance of the 3-D force sensors is improved, and the algorithm proves to be effective, practical, and robust in compensating for the sensors’ dynamic behavior.