Melisa Orta Martinez

Soft robots excel in adaptability, and rigid robots in precise, forceful actuation, yet each faces limitations that reduce performance in cluttered or constrained settings. We address these gaps by enabling physical coupling between a soft Vine robot and centimeter-scale RESCUE roller robots, allowing heterogeneous teams to work cooperatively to leverage each other’s strengths. Our approach improves mobility, enables power sharing, and communication through mechanical interfaces, modular chain formation, and high-speed recharging and data transfer via electrical contact using magnetic connectors. RESCUE rollers can form chains for collaborative locomotion and load distribution, and seamlessly integrate with the Vine robot to extend its workspace. In return, the Vine robot serves as a mobile power and communication hub, recharging rollers and relaying data to extend operational time. Experimental validation includes measurements of pushing and pulling strength in roller chains (Section III-A), Vine robot buckling under load (Section III-B), and demonstrations of Vine robot and RESCUE rollers interactions in a cluttered environment (Section IV-A), tight turning of the Vine robot using rollers (Section IV-B), and in-field recharging of RESCUE rollers (Section IV-C). This work advances cooperative strategies for heterogeneous robotic systems, enabling adaptable and resilient performance in challenging environments.

Vine robots offer unique capabilities for navigating confined environments through growth via eversion, making them suitable for urban search and rescue operations. Prior research has demonstrated the potential of vine robots to deploy passive payloads or mobile agents for distributed mapping and exploration. However, existing deployment methods rely on passive delivery, pushing payloads toward the robot tip during growth, thus limiting control over deployment timing and location. In this work, we introduce a compact wireless gate mechanism designed to achieve controlled payload deployment within the vine robot. This mechanism operates by selectively constricting the vine’s inner sleeve, securely retaining payloads and releasing them upon receiving a remote command. Fully integrated into the robot’s structure, the gate allows unimpeded vine growth and supports wireless communication with an external mobile controller. Experiments with benchtop and integrated vine robots show that the gate remains co-located with the tip and enables on-command hold and release without impeding eversion. The gate supports sequential delivery of multiple payloads and can deploy irregular payloads by preparing specific accommodations. These results establish timed, location-accurate deployment for USAR scenarios while keeping vine robot hardware composition regular.

Haptic devices have proven valuable in educational settings for enhancing science, technology, engineering, and mathematics (STEM) learning. By simulating touch-based interactions, these systems facilitate hands-on experiences that effectively convey complex engineering and mathematics concepts, positively influencing students' engagement in STEM subjects. However, to our knowledge, existing educational haptic devices have not specifically targeted teaching concepts related to dexterous robotic manipulation. In this paper, we introduce the HapDeltaZ, a novel, 3D-printed, low-cost, grounded kinesthetic haptic device featuring a compliant Delta mechanism. The device provides feedback in three degrees of freedom to a user's thumb, enabling interaction with virtual objects of varying stiffness. The HapDeltaZ design prioritizes optimizing the thumb's workspace to enhance dexterity and facilitate natural movement. We evaluated the performance of the HapDeltaZ through a user study, assessing participants' ability to differentiate virtual objects based on stiffness. Our results indicate that participants could consistently distinguish between different stiffness values across two virtual environments.

Exploring spatial aspects of digital interfaces remains challenging for people who are blind, despite advances in screen readers, particularly when these interfaces include graphical content. Previous research has demonstrated that absolute positioning combined with haptic feedback can significantly enhance performance in shape recognition tasks. However, most existing systems often rely on expensive graphical tablets, restricting their accessibility and scalability. In this work, we present advancements in a low-cost haptic mouse system by replacing the graphical tablet with an integration of an optical sensor and an inertial measurement unit (IMU). This combination enables driftcorrected absolute positioning, independent of device orientation. Additionally, we introduce a design employing linear resonance actuators to deliver directional vibrotactile cues to the user's palm, enhancing positional awareness and orientation during exploration. To evaluate the system, we performed controlled displacement tests to measure positioning accuracy, quantified as the deviation from expected trajectories (pixels per centimeter). Preliminary results demonstrate millimeter-scale resolution and consistent positional mapping across trials. This research contributes to accessible human-computer interaction by eliminating dependence on costly hardware for absolute positioning, providing a scalable solution suitable for educational applications (e.g., STEM-related graphical exploration) and general digital interface navigation for blind and visually impaired users. Future studies will involve user-centered evaluations with blind participants to further refine the haptic feedback design and validate the usability of the system.

This study investigates the conditions that foster effective interdisciplinary collaboration-defined as the integration of knowledge, practices, and perspectives from different disciplines-through qualitative analysis of project-based learning in higher education.Grounded in theories from the Learning Sciences and Computer-Supported Collaborative Learning, we analyze how knowledge transfer and leadership dynamics mediate interdisciplinary engagement in a course combining robotics, mathematics, and textile arts.Systematic analysis of qualitative data from 23 undergraduate students across six groups reveals three distinct patterns of collaborative engagement.Our findings show that successful collaboration hinges on (1) fluid leadership transitions that leverage diverse expertise and (2) structured opportunities for knowledge exchange.Groups with distributed leadership and consistent knowledge-sharing practices produced higher-quality projects than those with centralized leadership.These findings contribute to theoretical understandings of interdisciplinary learning and offer practical guidelines for designing collaborative learning environments in higher education.

This study explores how integrating computational thinking (CT) with weaving can make CT more accessible and engaging.Nineteen undergraduates participated in a rope weaving activity to recreate patterns, test structures, and connect math to tangible materials.Analysis of reflections and video data revealed students applying CT practices like pattern recognition and algorithmic thinking.Findings highlight the value of tactile, craft-based activities in supporting CT learning and bridging abstract concepts with real-world, embodied experiences.

This symposium focuses on understanding learners' interactions with artificial intelligence (AI), particularly chatbots and agents, to advance human-AI collaboration research from the perspective of computer-supported collaborative learning (CSCL).The four papers discuss human-AI interactions in various learning tasks and domains, especially from an augmentation perspective.The studies examine how these interactions can support learning processes and how learners perceive AI as an interaction partner (or not).By exploring these augmentation dynamics across tasks and domains, the symposium contributes to the evolving understanding of human-AI collaboration in teaching and learning.The symposium's discussant will synthesize insights from these studies to shed light on how these studies contribute knowledge and how these approaches need to be further developed to contribute to future hybrid intelligence (HI) systems in education.discussant Mutlu Cukurova (10-min).This will allow for a 15-minute general discussion between the symposium participants and the session attendees.At the beginning of their presentation, each presenter will define how their current research advances human-AI collaboration research from the perspective of computer-supported collaborative learning (CSCL) and contributes to the development of hybrid intelligence systems in education.

Urban search and rescue (USAR) remains a challenging application domain despite recent and rapid advancements in robotics. Navigating complex environments is difficult for robotic architectures due to the variation in surface composition and the range of terrain feature sizes. This variety of scenarios incentivize creating a range of different robot designs, but how to effectively use these robots together remains uncertain. Soft robots and centimeter (cm) scale robots have been studied for their potential uses in USAR operations due to their ability to pass through narrow spaces by deforming to adapt to the environment or by virtue of their small size. However, both classes still poses significant challenges in control, limited operational domain, and runtime. In this paper, we demonstrate how collaborative behavior of Vine Robots, i.e. soft inflatable growing robots, and RESCUE Rollers, i.e. cm-scale mobile robots, can expand the reachable area for each robot in rescue-like operations. We accomplish this expansion by equipping the Vine robot to carry and deploy the RESCUE Rollers, and by using the RESCUE Rollers to improve the maneuvarability of Vine Robots by external steering. Through experimental characterization of the necessary interaction forces and an example scenario of collaborative behavior to reach the top of a ramp, we demonstrate that this approach of physical heterogeneous collaboration can lead to increased capabilities of all systems involved.

A miniature haptic stimulation device utilizes focused ultrasound to deliver a tactile haptic sensation to the finger. The 1-3 piezocomposite device has a 1 cm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> footprint, which is an order of magnitude smaller than other ultrasonic haptic devices and is a good candidate for wearable tactile rendering systems. The device focuses energy to a 1 mm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> voxel. The current prototype was validated with a small, preliminary human subject study and requires an average input voltage of 68.8 V to elicit tactile sensation. The sensory drive voltage threshold will decrease with future refinement of mechanical impedance matching and focusing.

Weaving is a fabrication process that is grounded in mathematics and engineering: from the binary, matrix-like nature of the pattern drafts weavers have used for centuries, to the punch card programming of the first Jacquard looms. This intersection of disciplines provides an opportunity to ground abstract mathematical concepts in a concrete and embodied art, viewing this textile art through the lens of engineering. Currently, available looms are not optimized to take advantage of this opportunity to increase mathematics learning by providing hands-on interdisciplinary learning in collegiate classrooms. In this work, we present SPEERLoom: an open-source, robotic Jacquard loom kit designed to be a tool for interweaving cloth fabrication, mathematics, and engineering to support interdisciplinary learning in the classroom. We discuss the design requirements and subsequent design of SPEERLoom. We also present the results of a pilot study in a post-secondary class finding that SPEERLoom supports hands-on, interdisciplinary learning of math, engineering, and textiles.

Many academic subjects are inaccessible for students who are blind or have low vision (BLV) due to the prevalence of visual aids to represent concepts. Interactive devices offer promise as creative tools for teachers of the visually impaired (TVIs) as they can support real-time iteration of adapted learning materials, display changing information, and provide embodied learning experiences for BLV students. We conducted semi-structured interviews with 5 educators (all have TVI experience) and identified their considerations when creating adaptations, attitudes towards technology, and perspectives on existing barriers to access. Our findings reveal and reaffirm unresolved challenges in the adaptation process as well as offer insights into key factors that must be considered when selecting the type of adaptation. From these findings, we formulate design recommendations for interactive tools that support TVIs in creating effective adaptations for BLV students.

In this work, we introduce a novel haptic device composed of a wearable fingertip sheath, fabricated using an oleogel loaded with magnetic particles, and an external electromagnet. The sheath is actuated using the external magnetic field provided by the electromagnet, which is equipped with a field-focusing pole piece. The oleogel composite used in this device has been optimized for the transfer of the magnetic force from the material to the skin to provide perceptible forces to the wearer. We compare our composite to composites created with materials commonly used in the literature and find the force transfer from our material, as measured by a force sensor, to be much greater when actuated under the same range of input voltages to the electromagnet. We also present a psychophysical user study that shows a linear relationship between this range of input voltages and perceptual magnitude. This result indicates that the device provides a range of tactile feedback that can be driven to a desired intensity of sensation through proportional voltage control.