<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Projects on Yichen Andy Yu</title><link>https://yuyichen.net/projects/</link><description>Recent content in Projects on Yichen Andy Yu</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Sat, 10 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://yuyichen.net/projects/index.xml" rel="self" type="application/rss+xml"/><item><title>CollabPEAS</title><link>https://yuyichen.net/projects/collabpeas/</link><pubDate>Sat, 10 Oct 2026 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/collabpeas/</guid><description>&lt;p>CollabPEAS is a collaborative system for AI literacy in which pairs of learners come to understand how an AI agent&amp;rsquo;s behavior is produced by human specification. Working in shared virtual reality, learners jointly specify a virtual cleaning robot using the PEAS framework (Performance measure, Environment, Actuators, Sensors).&lt;/p>
&lt;p>The system combines three design elements: a five-step PEAS-guided design flow in VR that makes specification choices spatially grounded and observable; a block-based coding interface that preserves the perceive, decide, act reasoning cycle; and collaboration supports that structure negotiation and shared ownership between partners.&lt;/p></description></item><item><title>SpatialPrompt</title><link>https://yuyichen.net/projects/spatialprompt/</link><pubDate>Fri, 12 Jun 2026 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/spatialprompt/</guid><description>&lt;p>SpatialPrompt is an Extended Reality (XR) system that turns spatial sketches into executable constraints for controllable 3D generation. Users draw rough structures with a 3D pen and add voice prompts for semantic and stylistic intent.&lt;/p>
&lt;p>The system supports iterative refinement and synchronous co-creation in shared space, with color-coded contributions for each collaborator. It runs on Apple Vision Pro with Logitech Muse and Meshy. A heuristic evaluation suggests the workflow is intuitive and supports shared understanding in collaborative creation, while revealing needs for faster generation and clearer feedback.&lt;/p></description></item><item><title>RunPacer</title><link>https://yuyichen.net/projects/runpacer/</link><pubDate>Sun, 26 Oct 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/runpacer/</guid><description>&lt;p>Visually impaired runners often need a guide runner to run outdoors safely. Verbal cues and physical tethers can be mentally taxing and physically restrictive, and they reinforce dependency. RunPacer is a smartwatch-based vibrotactile feedback system that delivers synchronized rhythmic pulses to both runners, so the pair can coordinate their cadence intuitively with less cognitive load. It supports both fixed and adaptive cadence modes to promote shared agency and comfort.&lt;/p>
&lt;p>RunPacer began as an iOS app that won 1st Prize in the 2023 Apple iOS App Innovation Competition, with an invitation to WWDC 2023. Built with WatchKit and HealthKit, the app connects blind runners with running partners and tracks health metrics during each run.&lt;/p></description></item><item><title>NieNie</title><link>https://yuyichen.net/projects/nienie/</link><pubDate>Sun, 12 Oct 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/nienie/</guid><description>&lt;p>Young people face increasing psychological stress, and traditional stress management tools often rely on static scripts or passive content. NieNie combines rhythm biofeedback with real-time psychological guidance from a large language model, delivered through an interactive, tactile device.&lt;/p>
&lt;p>The system collects physiological signals such as heart rate variability and generates adaptive squeeze-release rhythms on a soft, squeezable controller, paired with Unity gameplay and supportive language prompts. By placing users in an embodied loop of touch, biofeedback and adaptive language support, NieNie connects bodily action with mental health in everyday contexts.&lt;/p></description></item><item><title>GenLARP</title><link>https://yuyichen.net/projects/genlarp/</link><pubDate>Wed, 01 Oct 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/genlarp/</guid><description>&lt;p>GenLARP is a virtual reality system that transforms personalized stories into immersive live action role-playing (LARP) experiences. Users act as both creators and players: they describe the story they want, design characters from their own descriptions, and then live in the story world.&lt;/p>
&lt;p>Generative AI and agents powered by large language models build the 3D scenes, give each role its profile, background and task, and drive the interactions between characters, so players can switch roles and understand a story by acting it out.&lt;/p></description></item><item><title>Noetic Dream</title><link>https://yuyichen.net/projects/noetic-dream/</link><pubDate>Sun, 28 Sep 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/noetic-dream/</guid><description>&lt;p>Lucid dreaming relies on a high level of metacognition, and induction techniques take significant time and effort to master. Noetic Dream is a personalized training system that combines virtual reality with open-monitoring (OM) meditation, acting on &amp;ldquo;dream awareness&amp;rdquo; through both external and internal pathways.&lt;/p>
&lt;p>VR provides immersive dream-based games in which users practice spotting unrealistic states, while OM meditation stabilizes internal focus and implants lucid intent. Large language models help reconstruct each user&amp;rsquo;s own dream scenes in VR, and multimodal cues across the training cycle help users build the habit of recognizing a dream.&lt;/p></description></item><item><title>MeloKids</title><link>https://yuyichen.net/projects/melokids/</link><pubDate>Thu, 24 Jul 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/melokids/</guid><description>&lt;p>Children with hearing impairments face ongoing challenges in language and motor development. MeloKids explores how multisensory feedback in virtual reality, integrating auditory, visual and tactile stimuli, can support their rehabilitation.&lt;/p>
&lt;p>In pitch-matching tasks, children see and feel their own voice: pitch is mapped in real time to a visual trajectory in VR and to a vibrotactile array on the forearm. Using functional near-infrared spectroscopy (fNIRS), we measure cortical activation across interaction modes to inform personalized, interactive rehabilitation systems that strengthen cognitive engagement and motor control.&lt;/p></description></item><item><title>Chameleon</title><link>https://yuyichen.net/projects/chameleon/</link><pubDate>Sat, 26 Apr 2025 00:00:00 +0000</pubDate><guid>https://yuyichen.net/projects/chameleon/</guid><description>&lt;p>In VR, free movement in real space enhances immersion but raises the risk of colliding with real-world obstacles. Prior work substitutes obstacles with context-related virtual objects but treats every obstacle the same, which can reduce awareness of high-risk obstacles and miss the chance to use low-risk ones for haptic feedback and interactivity.&lt;/p>
&lt;p>Chameleon classifies real-world obstacles by risk level and substitutes them with context-related virtual objects whose appearance matches each obstacle&amp;rsquo;s real-world risk, keeping users both safe and immersed. A preliminary heuristic evaluation assessed the usability of visual textures as an implicit signal of obstacle risk.&lt;/p></description></item></channel></rss>