Strengthening human-robot symbiosis through multimodal sensing

Daniel Freer, Benny Lo
<span title="2021-05-07">2021</span>
As robotic technology has advanced throughout the ages, one of its primary functions has been to assist humans in completing difficult or dangerous tasks. Through the lens of outer space and healthcare, this thesis explores the possibility for a symbiotic relationship between humans and assistive robots by improving training protocols and with enhanced sensor data processing. This relationship is explored through several technologies: 1) a telerobotic simulator for training; 2) a wearable
more &raquo; ... c sensing system for the arm; 3) motor imagery control of an assistive robot; and 4) autonomous five-fingered grasping. Within these platforms, various sensor types are used including electroencephalography (EEG), electromyography (EMG), tendon-based sensing, and computer vision (CV). At the intersection of these robotic technologies and sensors, the key results include a novel metric based on Riemannian geometry indicating new EEG mental workload features, and EMG analysis showing which preprocessing methods are most promising for feature selection. In addition, EEG motor imagery training protocols can be improved through novel data augmentation strategies which reduce the amount of needed training data per person, and an adaptive task-based protocol which could more readily transfer into the real world. Finally, CV-based robotic hand shaping was shown to be a promising way to achieve a robust grasp with a five-fingered gripper. The advancements made by studying each of these platforms and sensors provide a new link between humans and robots, which in turn creates more shared understanding between the biological and digital domains. It is only through this shared understanding that we can create a future where humans and robotic systems work together with efficiency and security.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25560/88276">doi:10.25560/88276</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/apglqtxnk5h2tkyjx7ijtyctoe">fatcat:apglqtxnk5h2tkyjx7ijtyctoe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210731230746/https://spiral.imperial.ac.uk:8443/bitstream/10044/1/88276/3/Freer-D-2021-PhD-Thesis.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f0/f4/f0f4e54550ddbcf3d995a3da6ee7e41491215576.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25560/88276"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> Publisher / doi.org </button> </a>