Air-Chamber Based Soft Six-Axis Force/Torque Sensor for Human-Robot Interaction

TL;DR

This paper tackles safe, high-coverage force/torque sensing for human–robot interaction by introducing a soft air-chamber based 6-F/T sensor composed of $16$ channels embedded in a two-layer rigid-soft structure. A model-based decoupling method converts the six-axis measurements into two independent three-axis problems, reducing calibration parameters from $96$ to $6$ and enabling practical wearability. Finite element analysis and comprehensive experiments confirm a sensing range of $50\ \mathrm{N}$ in force and $1\ \mathrm{Nm}$ in torque, with average metrics of $E_{dev}=4.9\%$, $E_{rep}=2.7\%$, $E_{nlin}=5.8\%$, and $E_{hys}=6.7\%$, and demonstrate a dynamic response with transfer functions around $G(s)\approx \frac{\text{gain}}{\tau s+1}$ and time constant $\bar{T}\approx 0.0034\ \mathrm{s}$. The work offers a soft yet accurate sensing solution suitable for wearable and human–robot interfaces, with clear pathways to improve permeability, reduce size, and extend bandwidth for real-time interaction.

Abstract

Soft multi-axis force/torque sensors provide safe and precise force interaction. Capturing the complete degree-of-freedom of force is imperative for accurate force measurement with six-axis force/torque sensors. However, cross-axis coupling can lead to calibration issues and decreased accuracy. In this instance, developing a soft and accurate six-axis sensor is a challenging task. In this paper, a soft air-chamber type six-axis force/torque sensor with 16-channel barometers is introduced, which housed in hyper-elastic air chambers made of silicone rubber. Additionally, an effective decoupling method is proposed, based on a rigid-soft hierarchical structure, which reduces the six-axis decoupling problem to two three-axis decoupling problems. Finite element model simulation and experiments demonstrate the compatibility of the proposed approach with reality. The prototype's sensing performance is quantitatively measured in terms of static load response, dynamic load response and dynamic response characteristic. It possesses a measuring range of 50 N force and 1 Nm torque, and the average deviation, repeatability, non-linearity and hysteresis are 4.9$\%$, 2.7$\%$, 5.8$\%$ and 6.7$\%$, respectively. The results indicate that the prototype exhibits satisfactory sensing performance while maintaining its softness due to the presence of soft air chambers.

Figures (12)

• Figure 1: The overview of the proposed sensor. (a) The proposed prototype model. (b) The front section view of the sensor model. (c) The explosion view of the sensor model. (d) The top section view of the upper layer sensor model. The numbers refer to the air chambers, $u$ donates the upper layer, $l$ donates the lower layer.
• Figure 2: Fabrication process of the proposed six-axis sensor. (a) Step 1, fabricate the upper soft layer. $\rm a_1$, 3D printed mold, pour the silicone rubber into the cavity of the molding. $\rm a_2$, the enlarged cavity mold. (b) Step 2, fabricate the lower hard layer. $\rm b_1$, hard layer mold. $\rm b_2$, pour liquid silicone rubber. $\rm b_3$, cover the lid. $\rm b_4$, embedded the barometers and seal the bottom. (c) The internal image with absolute barometers. (d) Step 3, use mold to merge the upper layer and the lower layer. (e) The internal images without barometers. $\rm e_1$, upper soft layer. $\rm e_2$, lower hard layer. (f) The prototype ($\rm f_1$) and the bendable characteristics display of the soft upper layer ($\rm f_2$).
• Figure 3: Cross-sectional area deformation analysis of pneumatic chambers. (a) Case I, applied single axis force $F_z$, the upper layer and the lower layer are all compressed. (b) Case II, applied single axis force $F_x/F_y$, the upper layer slips along with the direction of the force, the lower layer makes no change. (c) Case III, applied single axis torque $T_z$, the upper layer rotates along with z-axis (the dotted line shows the deformation), the lower layer makes no change. (d) Case IV, applied single axis torque $T_x/T_y$, one side is compressed and the other side is stretched.
• Figure 4: Cross-sectional area of the air chamber model analysis defined by $A_1(\theta)$. A cylindrical column at the center of each chamber is used to support the cavity and prevent excessive collapse.
• Figure 5: The single air chamber deformation analysis. Green area represents the central pillar and gray area refers to the air chamber. (a) Case I, $F_z$ applied. (b) Case II, $F_x/F_y$ applied. (c) Case III, $T_z$ applied. (d) Case IV, $T_x/T_y$ applied.