A Novel 6-axis Force/Torque Sensor Using Inductance Sensors

This paper presents a novel six-axis force/torque (F/T) sensor based on inductive sensing technology. Unlike conventional strain gauge-based sensors that require direct contact and external amplification, the proposed sensor utilizes non-contact inductive measurements to estimate force via displacement of a conductive target. A compact, fully integrated architecture is achieved by incorporating a CAN-FD based signal processing module directly onto the PCB, enabling high-speed data acquisition at up to 4~kHz without external DAQ systems. The sensing mechanism is modeled and calibrated through a rational function fitting approach, which demonstrated superior performance in terms of root mean square error (RMSE), coefficient of determination ($R^2$), and linearity error compared to other nonlinear models. Static and repeatability experiments validate the sensor's accuracy, achieving a resolution of 0.03~N and quantization levels exceeding 55,000 steps, surpassing that of commercial sensors. The sensor also exhibits low crosstalk, high sensitivity, and robust noise characteristics. Its performance and structure make it suitable for precision robotic applications, especially in scenarios where compactness, non-contact operation, and integrated processing are essential.

Temperature Compensation Method of Six-Axis Force/Torque Sensor Using Gated Recurrent Unit

This study aims to enhance the accuracy of a six-axis force/torque sensor compared to existing approaches that utilize Multi-Layer Perceptron (MLP) and the Least Square Method. The sensor used in this study is based on a photo-coupler and operates with infrared light, making it susceptible to dark current effects, which cause drift due to temperature variations. Additionally, the sensor is compact and lightweight (45g), resulting in a low thermal capacity. Consequently, even small amounts of heat can induce rapid temperature changes, affecting the sensor's performance in real time. To address these challenges, this study compares the conventional MLP approach with the proposed Gated Recurrent Unit (GRU)-based method. Experimental results demonstrate that the GRU approach, leveraging sequential data, achieves superior performance.

Parameter Optimization of Optical Six-Axis Force/Torque Sensor for Legged Robots

This paper introduces a novel six-axis force/torque sensor tailored for compact and lightweight legged robots. Unlike traditional strain gauge-based sensors, the proposed non-contact design employs photocouplers, enhancing resistance to physical impacts and reducing damage risk. This approach simplifies manufacturing, lowers costs, and meets the demands of legged robots by combining small size, light weight, and a wide force measurement range. A methodology for optimizing sensor parameters is also presented, focusing on maximizing sensitivity and minimizing error. Precise modeling and analysis of objective functions enabled the derivation of optimal design parameters. The sensor's performance was validated through extensive testing and integration into quadruped robots, demonstrating alignment with theoretical modeling. The sensor's precise measurement capabilities make it suitable for diverse robotic environments, particularly in analyzing interactions between robot feet and the ground. This innovation addresses existing sensor limitations while contributing to advancements in robotics and sensor technology, paving the way for future applications in robotic systems.

A Compact Optical Six-Axis Force/Torque Sensor for Legged Robots Using a Polymorphic Calibration Method

This paper presents a novel design for a compact, lightweight 6-axis force/torque sensor intended for use in legged robots. The design promotes easy manufacturing and cost reduction, while introducing innovative calibration methods that simplify the calibration process and minimize effort. The sensor's advantages are achieved by streamlining the structure for durability, implementing noncontact sensors, and providing a wider sensing range compared to commercial sensors. To maintain a simple structure, the paper proposes a force sensing scheme using photocouplers where the sensing elements are aligned in-plane. This strategy enables all sensing elements to be fabricated on a single printed circuit board, eliminating manual labor tasks such as bonding and coating the sensing elements. The prototype sensor contains only four parts, costs less than $250, and exhibits high response frequency and performance. Traditional calibration methods present challenges, such as the need for specialized equipment and extensive labor. To facilitate easy calibration without the need for specialized equipment, a new method using optimal control is proposed. To verify the feasibility of these ideas, a prototype six-axis F/T sensor was manufactured. Its performance was evaluated and compared to a reference F/T sensor and previous calibration methods.