Abstract
To enable precise manipulation of fragile objects, robots need the sense of touch. Localized sensing of the 3D force vector with a few mN of resolution is desired. Magnetic force sensors, consisting of a magnet embedded in an elastomer and a 3D magnetometer, were demonstrated as desirable candidates. These prior-art sensors measure the local 3D magnetic field at a single sensing pixel. Hence, they cannot distinguish between the signal and stray magnetic fields. Any stray field directly leaks into the force readout signal path. This letter introduces a design immune to stray magnetic fields. The sensor uses multiple magnetic pixels, and operates on the gradient of the magnetic field. The pixels and conditioning electronics are fully integrated on chip. The 3D force calculation is based on a polynomial model fitted to a calibration data set. The impact of a 2-mT magnetic stray field on the force output is limited to 0.3% of the full scale—about two orders magnitude improvement over the prior art. The force resolution is 2.7 mN and remains competitive. Furthermore, an on-chip temperature sensor and an algorithm are used to compensate the intrinsic thermal drift in the range 0–50 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">${}^{\circ }$</tex-math></inline-formula> C. To validate the proper force regulation in spite of a nearby magnet, we integrated our prototype into a robotic hand. Our results demonstrate the robustness of 3D magnetic force sensors in the presence of real-world parasitic disturbances.
Highlights
This demonstrated the unique advantages of the magnetic force sensors: 3D force sensing in a compact and cost-effective package with just one chip per tactile pixel (”taxel” in the literature: TActile piXEL)
The weight matrix was trained on this data set, optimizing its components based on a stochastic gradient descent algorithm with the Huber loss function and a L1 regularization term
The recent magnetic force sensor [29] achieves 3D force vector sensing, state-of-the-art resolution and compactness, but the sensitivity to stray fields remains a key limitation
Summary
G IVING robots the sense of touch is a key challenge in robotics to enable manipulation of objects with human levels of dexterity [1]. Robots require tactile sensors providing measurements of the normal and tangential forces at the points of contact [2]. The sensor contact surface should be compliant to increase the contact area, and have a skin-like feeling. The 6D force/torque sensors in [6] and [7] based on strain gauges and variable capacitors respectively. Such sensors are better suited for measuring the force and torque at the joints (as opposed to the contact at the skin surface)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
