Optimization Design of Robot Joint Torque Sensor With Y-type Beam Structure

Joint torque sensor (JTS) is the key to achieving human-machine collaboration in robots. This article proposes the JTS based on the Y-type elastic beam structure. A multi-objective optimization equation is established with sensitivity, radial stiffness, and axial stiffness as performance indicators, and the optimal structural parameters are obtained by the Response Surface Method (RSM) based on Box-Behnken experimental design and multi-objective genetic algorithm (MOGA). After simulation experiments, the optimized axial stiffness and radial stiffness of the JTS are $4427.268 \mathrm{Nm} / \mathrm{rad}$ and $2060.397 \mathrm{Nm} / \mathrm{rad}$, respectively, and the working area average equivalent elastic strain is $1.018 \mu \mathrm{~m} / \mathrm{mm}$. Compared with the original size of the JTS, the average equivalent elastic strain, radial stiffness, and axial stiffness are improved by $3.041 \%, 9.092 \%$, and $9.471 \%$, respectively. The range is $0 \sim \pm 100 \mathrm{Nm}$, and the overload resistance is $300 \%$ F.S. In summary, the sensor performance meets engineering requirements and guides subsequent sensor production and manufacturing.