Surface ripple suppression in subaperture polishing with fragment-type tool paths.

In order to reduce the ripple-induced effect in subaperture polishing of high-power laser optics, the tool paths are investigated and optimized in this study. Various fragment-type curves (i.e., fractal and fractal-like unicursal curves) are designed as tool paths, and most importantly, a novel algorithm for generating a random fractal-like tool path is proposed in detail. Compared with scanning paths, fragment-type paths possess multi-directionality during multiple polishing iterations, which could better smooth the surface texture and restrain the polishing-induced surface ripples. In particular, the random fractal-like path exhibits high randomness, boundary adaption, and step-length arbitrariness, which make it more flexible and powerful for iterative polishing. Experiments were conducted on a computer numerical control machine, with conclusions that during iterative polishing, (1) the repetitive usage of a single fractal path would aggravate surface ripples; (2) the combination of multiple fractal paths or using the random fractal-like path can restrain surface ripples efficiently; and (3) pitch pads exhibit a better ripple-smoothing effect than polyurethane. The proposed random fractal-like path is a good prospect for the manufacture of high-power laser optics, which demands very low surface ripples.

Accurate raycasting selection with rotation gesture using a 6-DOF tracking device.

A six degrees of freedom (6-DOF) controller is a commonly used input device in three-dimensional user interface (3DUI) applications. However, for the fundamental task of target selection in 3D space, the selection accuracy decreases owing to the Heisenberg effect during the manipulation of the 6-DOF controller. Based on the pointing action of a 6-DOF device, we establish the mathematical model for raycasting and analyze the possibility of using a rotation gesture to select a target. This study proposes a target selection method using the axial rotation of the user's wrist, which reduces the negative impact of a discrete input for triggering the selection on accuracy. The detection model can identify the start time of the user's rotation action. The custom designed control display gain (CD gain) function can maintain the ray's stability during the rotation gesture. This method was verified using a 6-DOF pen and Vive controller in three experiments. The results show that the accuracy of the proposed rotation gesture-based raycasting target selection method is superior to that of the traditional button-pressing method, and it can be integrated into existing tracking systems for further application.

Vibrotactile Feedback Rendering of Patterned Textures Using a Waveform Segment Table Method.

Vibrotactile feedback is a common form of rendered haptic feedback used for simulating stylus-texture interaction. Most state-of-the-art stylus-texture interaction vibrotactile feedback synthesizing methods are oriented toward generating signal with resemblance in spectrum in frequency domain. In this paper we set our foot backward and explore more about record-and-playback method for a subset of textures: those that have obvious spatial pattern, which constitutes a significant proportion of man-made textures we interact with in daily life. We propose a method that explicitly renders the periodic vibrotactile feedback for patterned textures. The method uses Dynamic Time Warping to select the most representative signal segment from a long continuous signal captured under a certain interaction condition, and constructs a waveform segment table to store representative signal segments under different conditions. Results of similarity-comparison user study show that subjects gave generally higher similarity scores to our proposed method than to a spectrum-oriented method. The results shed light on the importance of conserving the pattern in the haptic feedback rendering for patterned textures.

Estimating Forearm Axial Rotation Using Vive Trackers for Interaction With Serious Games.

One of the most common upper limb dysfunction seen among cerebral palsy children is the limited supination or pronation of the forearm, thus training forearm axial rotation is a common demand for interventions. We propose a method to estimate forearm axial rotation using commercially available tracking device HTC Vive tracker to support interaction with serious games with forearm axial rotation movement for rehabilitation in clinic or home environment. The proposed method provides an easy-to-access and affordable way for forearm axial angle estimation when arm is at different positions relative to torso and when tracking devices are worn in different orientations. Effectiveness of the proposed method is examined, and a serious game system that uses the method for interaction is designed which was installed in The First Hospital of Jilin University as an option of forearm movement practices for children with cerebral palsy.