Researchers from Florida Atlantic University’s College of Engineering and Computer Science created photolithographically manufactured, highly robust liquid metal tactile sensors (LMS) and successfully incorporated them into a prosthetic hand’s fingertips, helping to restore tactile sensation with prosthetic limb use.
Devising a way to integrate hierarchical multi-finger tactile sensations has become a challenge of great interest for the field of biomedical research, as it could provide prosthetic hands with a greater level of intelligence.
Touch sensation for prosthetic hands is required to improve the experience of upper limb amputees in everyday activities. Because the amputee is not directly aware of the prosthetic fingertip forces when the afferent brain connection is destroyed, the lack of sensory feedback can be a trying issue such as when things being grasped are unintentionally crushed. Tactile sensors for prosthetic hands have received a lot of attention, but advancements in lightweight, low-cost, and robust multimodal tactile sensors are still needed.
In a new study, researchers from Florida Atlantic University’s College of Engineering and Computer Science developed highly flexible liquid metal tactile sensors (LMS), which were photolithographically produced, and successfully integrated them into a prosthetic hand’s fingertips.
The paper states, “hierarchical tactile sensation integration from multiple fingers is a trait exhibited by people, and could be a useful technique for a haptic display to improve prosthetic hand functionality or to augment the intelligence of autonomous manipulators.”
Newly developed commercial prosthetics demonstrate a trend toward improved dexterity, although they lack tactile sensory skills while interacting with the environment and handling objects. Human hand control systems rely heavily on these touch sensations for handling objects; yet, people who have had their upper limbs amputated lack tactile sensations.
Photolithography was used to microfabricate the LMS mold. The LMS microchannel cross-section with dimensions of 400 m x 100 m was designed using SolidWorks. The LMS was integrated into the fingertips of the i-limb Ultra prosthetic hand using distal phalanx support components.
The i-limb’s sensorless fingertips were removed, and the new LMS-equipped fingertips were attached to the limb using the same connection points. Four alternative textures were constructed with one variable parameter- the distance between the ridges.
With individual fingers, the LMSs were able to successfully discern between varied sliding contact speeds and textures. The ability to discern between ten complex multi-textured surfaces using hierarchical tactile sensation integration from four fingertips at the same time was also exhibited. The findings of the paper suggest that these novel LMS applications to robotic hands could lead to further advancements in prosthetic hand functionality and haptic feedback.
The paper concluded by stating, “due to the compliant, lightweight nature of the LMS and high classification accuracy, this paper has demonstrated the feasibility of their application to robotic hands”.
The new research findings were published in Sensors, on June 24th, 2021.
Abstract. Multifunctional flexible tactile sensors could be useful to improve the control of prosthetic hands. To that end, highly stretchable liquid metal tactile sensors (LMS) were designed, manufactured via photolithography, and incorporated into the fingertips of a prosthetic hand. Three novel contributions were made with the LMS. First, individual fingertips were used to distinguish between different speeds of sliding contact with different surfaces. Second, differences in surface textures were reliably detected during sliding contact. Third, the capacity for hierarchical tactile sensor integration was demonstrated by using four LMS signals simultaneously to distinguish between ten complex multi-textured surfaces. Four different machine learning algorithms were compared for their successful classification capabilities: K-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and neural network (NN). The time-frequency features of the LMSs were extracted to train and test the machine learning algorithms. The NN generally performed the best at the speed and texture detection with a single finger and had a 99.2 ± 0.8% accuracy to distinguish between ten different multi-textured surfaces using four LMSs from four fingers simultaneously. The capability for hierarchical multi-finger tactile sensation integration could be useful to provide a higher level of intelligence for artificial hands.
Abd MA, Paul R, Aravelli A, Bai O, Lagos L, Lin M, Engeberg ED. Hierarchical Tactile Sensation Integration from Prosthetic Fingertips Enables Multi-Texture Surface Recognition. Sensors. 2021; 21(13):4324. https://doi.org/10.3390/s2113432