Wearable supernumerary robotic limbs (SRLs) sit at the intersection of human augmentation and embodied AI, promising to function as extensions of the human body. However, their movements within the intimate near-body space raise unresolved challenges for perceived safety, user control, and trust. In this paper, we present results from a Wizard-of-Oz study (n=18), where participants completed near-body collaboration tasks with SRLs to explore these challenges. We collected qualitative data through think-aloud protocols and semi-structured interviews, complemented by physiological signals and post-task ratings. Findings indicate that greater autonomy did not inherently enhance perceived safety or trust. Instead, participants identified near-body zones and paired them with clear coordination rules. They also expressed expectations for how different arm components should behave, shaping preferences around autonomy, perceived safety, and trust. Building on these insights, we introduce SRL Proxemics, a zone- and segment-level design framework showing that autonomy is not monolithic: perceived safety hinges on spatially calibrated, legible behaviors, not on autonomy level alone.

Cooling feedback has been shown to improve user experience in virtual reality. However, many existing methods face limitations, such as requiring skin contact, relying on airflow, having bulky components, or delivering unstable sensations. These issues limit their applicability to wearable systems, especially in scenarios demanding high responsiveness and perceptual consistency. We propose ChillFactor, a highly responsive, non-contact cooling interface for head-mounted displays (HMDs). It uses ultrasonic atomization of a predetermined alcohol solution to generate evaporative cooling near the skin, reducing airflow-induced artifacts and providing relatively consistent cold sensations. To validate the design, we first evaluated the system's cooling performance. At maximum output, it lowered skin temperature by approximately 2.6 °C within five seconds, a rate sufficient to induce an almost immediate perception of cold. We subsequently integrated ChillFactor into an HMD and conducted a user study with 12 participants in immersive VR scenarios. Participants reported perceiving the cold within 0.66 seconds and noted significantly higher levels of presence and realism compared to conditions without cooling feedback, demonstrating the system's effectiveness in enhancing immersion through rapid and adjustable thermal cues. This work provides a practical foundation for non-contact cooling feedback and expands the design space for multisensory interaction in virtual environments.

Real-world impact, such as hitting a tennis ball and a baseball, generates instantaneous, directional impact forces. However, current ungrounded force feedback technologies, such as air jets and propellers, can only generate directional impulses that are 10x-10,000x weaker. We present SpinShot, a flywheel-based device with a solenoid-actuated stopper capable of generating directional impulse of 22Nm in 1ms, which is more than 10x stronger than prior ungrounded directional technologies. Furthermore, we present a novel force design that reverses the flywheel immediately after the initial impact, to significantly increase the perceived magnitude. We conducted a series of two formative, perceptual studies (n=16, 18), followed by a summative user experience study (n=16) that compared SpinShot vs. moving mass (solenoid) and vs. air jets in a VR baseball hitting game. Results showed that SpinShot significantly improved realism, immersion, magnitude (p < .01) compared to both baselines, but significantly reduced comfort vs. air jets primarily due to the 2.9x device weight. Overall, SpinShot was preferred by 63-75% of the participants.

Impact events, which generate directional forces with extremely short impulse durations and large force magnitudes, are prevalent in both virtual reality (VR) games and real-world experiences. However, despite recent advancement in ungrounded force feedback technologies, such as air jet propulsion and propellers, these technologies remain 5-100x weaker and 10-500x slower compared to real-world impact events. For instance, they can only achieve 4N with a minimal duration of 50-500ms compared to the 20-400N forces generated within 1-5ms for baseball, ping-pong, drumming, and tennis. To overcome these limitations, we present AirCharge, a novel haptic device that accumulates air propulsion momentum to generate instantaneous, directional impact forces. By mounting compressed air jets on rotating swingarms, AirCharge can amplify impact force magnitude by more than 10x while matching real-world impulse duration of 3ms. To support high-frequency impacts, we explored and evaluated a series of device designs, culminating in a novel reciprocating dual-swingarm design that leverages a reversing bevel gearbox to eliminate gyro effects and to achieve impact feedback of up to 10Hz. User experience evaluation (n = 16) showed that AirCharge significantly enhanced realism and is preferred by participants compared to air jets without the charging mechanism.

We present DrivingVibe, which explores vibrotactile feedback designs around the head to enhance VR driving motion experiences. We propose two approaches that use a 360-degree vibrotactile headband: 1) mirroring and 2) 3D inertia-based. The mirroring approach extends the vibrotactile patterns of handheld controllers to actuate the entire headband uniformly. The 3D inertia-based approach uses the acceleration telemetry data that driving games/simulators export to motion platforms to generate directional vibration patterns, including: i) centrifugal forces, ii) horizontal acceleration/deceleration, and iii) vertical motion due to rough terrain. The two approaches are complementary as the mirroring approach supports all driving games because it does not require telemetry data, while the 3D inertia-based approach provides higher feedback fidelity for games that provide such data. We conducted a 24-person user experience evaluation in both passive passenger mode and active driving mode. Study results showed that both DrivingVibe designs significantly improved realism, immersion, and enjoyment (p<.01) with large effect sizes for the VR driving experiences. For overall preference, 88% (21/24) of participants preferred DrivingVibe, with a 2:1 preference for 3D inertia-based vs. mirroring designs (14 vs. 7 participants). For immersion and enjoyment, 96% (23/24) of participants preferred DrivingVibe, with nearly a 3:1 preference (17 vs. 6 participants) for the 3D inertia-based design.