The use of machine learning in embedded systems is an interesting topic, especially with the growth in popularity of the Internet of Things (IoT). The capacity of a system, such as a robot, to self-localize, is a fundamental skill for its navigation and decision-making processes. This work focuses on the feasibility of using machine learning in a Raspberry Pi 4 Model B, solving the localization problem using images and fiducial markers (ArUco markers) in the context of the RobotAtFactory 4.0 competition. The approaches were validated using a realistically simulated scenario. Three algorithms were tested, and all were shown to be a good solution for a limited amount of data. Results also show that when the amount of data grows, only Multi-Layer Perception (MLP) is feasible for the embedded application due to the required training time and the resulting size of the model.
The ever-increasing electrification of society has been a cause of utility grid issues in many regions around the world. With the increased adoption of electric vehicles (EVs) in the Netherlands, many new charge points (CPs) are required. A common installation practice of CPs is to group multiple CPs together on a single grid connection, the so-called charging hub. To further ensure EVs are adequately charged, various control strategies can be employed, or a stationary battery can be connected to this network. A pilot project in Amsterdam was used as a case study to validate the Python model developed in this study using the measured data. This paper presents an optimisation of the battery energy storage capacity and the grid connection capacity for such a P&R-based charging hub with various load profiles and various battery system costs. A variety of battery control strategies were simulated using both the optimal system sizing and the case study sizing. A recommendation for a control strategy is proposed.
Introducing a hyperbolic vortex into a showerhead is a possibility to achieve higher spray velocities for a given discharge without reducing the nozzle diameter. Due to the introduction of air bubbles into the water by the vortex, the spray is pushed from a transition (dripping faucet) regime into a jetting regime, which results in higher droplet and jet velocities using the same nozzle diameter and throughput. The same droplet and jet diameters were realized compared to a showerhead without a vortex. Assuming that the satisfaction of a shower experience is largely dependent on the droplet size and velocity, the implementation of a vortex in the showerhead could provide the same shower experience with 14% less water consumption compared to the normal showerhead. A full optical and physical analysis was presented, and the important chemical parameters were investigated.
MULTIFILE
