In this paper, we investigate the problem of grasping previously unseen objects in unstructured environments which are cluttered with multiple objects. Object geometry, reachability, and force-closure analysis are considered to address this problem. A framework is proposed for grasping unknown objects by localizing contact regions on the contours formed by a set of depth edges generated from a single-view 2D depth image. Specifically, contact regions are determined based on edge geometric features derived from analysis of the depth map data. Finally, the performance of the approach is successfully validated by applying it to scenes with both single and multiple objects, in both simulation and experiments. Using sequential processing in MATLAB running on a 4th-generation Intel Core Desktop, simulation results with the benchmark Object Segmentation Database show that the algorithm takes 281 ms on average to generate the 6D robot pose needed to attach with a pair of viable grasping edges that satisfy reachability and force-closure conditions. Experimental results in the Assistive Robotics Laboratory at UCF using a Kinect One sensor and a Baxter manipulator outfitted with a standard parallel gripper showcase the feasibility of the approach in grasping previously unseen objects from uncontrived multi-object settings.

In this paper an RFID-based localization algorithm for Internet of Things (IoT) is proposed. The responses of RFID tags to different readers are employed to determine the location of the IoT devices. The proposed algorithm does not rely on the on the unreliable power level reading and the sophisticated directionof-arrival-based methods. First, we assume that the location of the RFID readers are known a priori. Later, in this paper, we extend our results to the scenario where neither the location of IoT devices nor the RFID readers are known. In addition, we provide analytical bounds on the number of readers in the network to have a reliable localization results. The results of the numerical experiment show a highly accurate localization of IoT devices when our proposed RF-Localize algorithm is employed.

The goal of assistive robotic devices, such as a wheelchair-mounted robotic arms (WMRA), is to increase users’ functional independence. At odds with this goal is the fact that device interfaces tend to be rigid, requiring the user to adapt, rather than adapting to the user. Paperno, et al. (2016) identified key physical, cognitive, and sensory capabilities that affect an individual’s performance of simulated activities of daily living (e.g. picking up an object from the floor) while using a WMRA. Greater visual abilities (visual acuity, contrast sensitivity, and depth perception), cognitive abilities (processing speed, working memory, and spatial ability) and physical abilities (dexterity) resulted in participants completing tasks more quickly and with fewer total moves. We propose that interfaces should adapt to compensate for deficits in these capabilities to support a wider range of users. A variety of compensations should be developed and tested in order to identify the most effective techniques. For instance, object segmentation, a computer vision technique that separates objects and background in a visual scene, may compensate for deficits in contrast sensitivity, depth perception, processing speed, and working memories. However, contrast sensitivity may be better compensated for by use of a simple yellow filter on the screen, mimicking yellow lenses in glasses used for the same purpose. Similarly, depth perception limitations may be better overcome through the use of multiple camera views or by automating the pick-up and release mechanisms of the gripper. Thus there may be one compensation that facilitates WMRA use for a multitude of decrements or each factor may be better served by a specific separate compensation. In incorporating the effective compensations into the interface software, there should also be a capability of identifying which specific compensations should be activated for an individual user. For this we propose testing for these important individual differences should be included within the software. Virtual or online testing already exist for many of the identified factors and can be modified to fit our purpose. This is especially the case if gamification principles are applied as testing will engage user interest. In this way, the software can adjust compensations as a user’s visual, cognitive, and physical abilities change over time. Future research ventures will include identifying the most beneficial compensation for each identified individual difference and developing virtual gamified measures for those individual differences.

Motivated by grasp planning applications within cluttered environments, this paper presents a novel approach to performing real-time surface segmentations of never-before-seen objects scattered across a given scene. This approach utilizes an input 2D depth map, where a first principles-based algorithm is utilized to exploit the fact that continuous surfaces are bounded by contours of high gradient. From these regions, the associated object surfaces can be isolated and further adapted for grasp planning. This paper also provides details for extracting the six-DOF pose for an isolated surface and presents the case of leveraging such a pose to execute planar grasping to achieve both force and torque closure. As a consequence of the highly parallel software implementation, the algorithm is shown to outperform prior approaches across all notable metrics and is also shown to be invariant to object rotation, scale, orientation relative to other objects, clutter, and varying degree of noise. This allows for a robust set of operations that could be applied to many areas of robotics research. The algorithm is faster than real time in the sense that it is nearly two times faster than the sensor rate of 30 fps.

No abstract