A Temperature Compensation Method for aSix-Axis Force/Torque Sensor Utilizing Ensemble hWOA-LSSVM Based on Improved Trimmed Bagging

Abstract

The performance of a six-axis force/torque sensor (F/T sensor) severely decreased when working in an extreme environment due to its sensitivity to ambient temperature. This paper puts forward an ensemble temperature compensation method based on the whale optimization algorithm (WOA) tuning the least-square support vector machine (LSSVM) and trimmed bagging. To be specific, the stimulated annealing algorithm (SA) was hybridized to the WOA to solve the local entrapment problem, and an adaptive trimming strategy is proposed to obtain the optimal trim portion for the trimmed bagging. In addition, inverse quote error (invQE) and cross-validation are employed to estimate the fitness better in training process. The maximum absolute measurement error caused by temperature decreased from 3.34% to 3.9×10−3% of full scale after being compensated by the proposed method. The analyses of experiments illustrate the ensemble hWOA-LSSVM based on improved trimmed bagging improves the precision and stability of F/T sensors and possesses the strengths of local search ability and better adaptability.

Temperature Compensation of Wind Tunnel Balance Signal Detection System Based on IGWO-ELM

“…When operating under extreme conditions, the wide range of ambient temperature can induce significant drift in the system’s characteristics. The characteristic drifts will cause bias between measured values and real values, thereby affecting measurement accuracy [ 4 ]. Therefore, compensating for the temperature effect on the balance signal detection system is crucial, as the characteristic drift caused by temperature profoundly affects the accuracy of balance testing.…”

“…Initialization of gray wolf populations using circle mapping to increase the population diversity of the gray wolf algorithm. 4.…”

Temperature Compensation for Semiconductor Gas Sensors Based on Whale Optimization Algorithm–Least-Squares Support Vector Machine

“…(15,16) Software compensation usually uses information fusion techniques to build algorithm models to eliminate the effect of temperature on the sensor, which is more accurate and easier to implement than hardware compensation methods and is more commonly used for sensor temperature correction. (17) Li et al (18) addressed the problem of the serious degradation of the measurement accuracy of a six-axis force/torque sensor in extreme environments. They used a composite algorithm combining the simulated annealing algorithm (SAA) with the whale optimization algorithm (WOA) to optimize the least-squares support vector machine (LSSVM), which could effectively eliminate the interference of the temperature with the sensor output signal.…”