The exponentially growing Internet of Things (IoT) market presents compelling opportunities for innovators to develop and create meaningful applications affecting everyday life. This shifting IoT paradigm is placing new demands on educational institutions to train students with IoT-relevant skills. Numerous education researchers continue to document the implementation of IoT-based curriculum, pedagogy, and assessment techniques for STEM education. In this paper, we systematically reviewed this literature and identified 60 journal articles and conference papers that have reported implementing IoT curriculum and associated instructional approaches, educational technologies, and assessment strategies for K-12 and university students. The curriculum, pedagogy, and assessment strategies presented in these studies were analyzed in the context of the sensing, networking, services, and interface layers of the IoT technology paradigm. The review identifies best educational practices and synthesizes actionable strategies for educators to implement effective IoT learning experiences for students. These strategies leverage low-cost IoT hardware, open-source IoT software, active learning-based instructional approaches, and direct/indirect assessment methods. As IoT education becomes increasingly pervasive, the review and strategies provided in this paper may serve as a guide for future educational efforts.
This paper presents a real-time method to detect and track multiple mobile ground robots using event cameras. The method uses density-based spatial clustering of applications with noise (DBSCAN) to detect the robots and a single k-dimensional (k - d) tree to accurately keep track of them as they move in an indoor arena. Robust detections and tracks are maintained in the face of event camera noise and lack of events (due to robots moving slowly or stopping). An off-the-shelf RGB camera-based tracking system was used to provide ground truth. Experiments including up to 4 robots are performed to study the effect of i) varying DBSCAN parameters, ii) the event accumulation time, iii) the number of robots in the arena, iv) the speed of the robots, v) variation in ambient light conditions on the detection and tracking performance, and vi) the effect of alternative clustering algorithms on detection performance. The experimental results showed 100% detection and tracking fidelity in the face of event camera noise and robots stopping for tests involving up to 3 robots (and upwards of 93% for 4 robots). When the lighting conditions were varied, a graceful degradation in detection and tracking fidelity was observed.
The sustained growth of the solar energy industry has created the need to formalize solar energy education. This paper elucidates a novel, cloud-based, virtual reality (VR) pedagogical system that provides students with the fundamentals of solar or Photovoltaic (PV) cells, solar (PV) modules, and solar (PV) array installation configurations. The VR system was developed using Unity3d software and was integrated with a Learning Management System (LMS). The system was hosted on the Google Cloud Platform. Students/users accessed the VR system using a standard web browser, making it widely accessible. The VR system consisted of self-guided laboratory modules covering electrical engineering fundamentals of solar (PV) cells such as output power losses as a function of finger length, width, depth, and spacing. Additionally, series and parallel solar (PV) cell connections to create desired voltage and current output, and orientation/tilt considerations of solar (PV) array installations were covered. Physics models to incorporate realistic solar energy behavior were programmed in the VR system. Live graphs showing how parameters affected total power output provided instant feedback to students. An adaptive formative assessment system that tested students’ application of electrical engineering fundamentals along with design skills was implemented. The system was used in a first year engineering fundamentals undergraduate class to make solar energy education accessible to early state engineering students. Data collected from the cloud system and student survey indicate growth in student engagement and students’ knowledge of introductory solar and electrical engineering topics.
Motion capture systems are crucial in developing multi-quadrotor systems due to their ability to provide fast and accurate ground truth measurements for tracking and control. This paper presents the implementation details and experimental validation of a relatively low-cost motion-capture system for multi-quadrotor motion planning using an event camera. The real-time, multi-quadrotor detection and tracking tasks are performed using a deep learning network You-Only-Look-Once (YOLOv5) and a k-dimensional (k-d) tree, respectively. An optimization-based decentralized motion planning algorithm is implemented to demonstrate the effectiveness of this motion capture system. Extensive experimental evaluations were performed to (1) compare the performance of four deep-learning algorithms for high-speed multi-quadrotor detection on event-based data, (2) study precision, recall, and F1 scores as functions of lighting conditions and camera motion, and (3) investigate the scalability of this system as a function of the number of quadrotors flying in the arena. Comparative analysis of the deep learning algorithms on a consumer-grade GPU demonstrates a 4.8× to 12× sampling/inference rate advantage that YOLOv5 provides over representative one- and two-stage detectors and a 1.14× advantage over YOLOv4. In terms of precision and recall, YOLOv5 performed 15% to 18% and 27% to 41% better than representative state-of-the-art deep learning networks. Graceful detection and tracking performance degradation was observed in the face of progressively darker ambient light conditions. Despite severe camera motion, YOLOv5 precision and recall values of 94% and 98% were achieved, respectively. Finally, experiments involving up to six indoor quadrotors demonstrated the scalability of this approach. This paper also presents the first open-source event camera dataset in the literature, featuring over 10,000 fully annotated images of multiple quadrotors operating in indoor and outdoor environments.
12 January 2015 in IEEE Journal of Oceanic Engineering
Mixed integer nonlinear programming (MINLP) techniques are increasingly used to address challenging problems in robotics, especially multivehicle motion planning (MVMP). The main contribution of this paper is a discrete time, distributed receding horizon mixed integer nonlinear programming (RH-MINLP) formulation of the underwater multivehicle path coordination problem with constraints on vehicle kinematics, dynamics, collision avoidance, and acoustic communication connectivity, and the application of state-of-the-art MINLP solution techniques. Each vehicle robot starts from a fixed start point and moves toward a goal point along a fixed path, so as to avoid collisions and remain in communication connectivity with other robots. Acoustic communication connectivity constraints account for the attenuation due to signal propagation and delays arising from multipath propagation in noisy communication environments, and specify intervehicle connectivity in terms of a signal-to-noise ratio (SNR) threshold. Scenarios including up to four robots are simulated to demonstrate: 1) the effect of communication connectivity requirements on robot velocity profiles; and 2) the dependence of the solution computation time on the communication connectivity requirement. Typically the optimization improved connectivity at no appreciable cost in journey time (as measured by the arrival time of the last arriving robot). Results also demonstrate the responsive nature of robot trajectories to safety requirements with collision avoidance being achieved at all times despite overlapping and intersecting paths.