In this paper, we present a fast monocular depth estimation method for enabling 3D perception capabilities of low-cost underwater robots. We formulate a novel end-to-end deep visual learning pipeline named UDepth, which incorporates domain knowledge of image formation characteristics of natural underwater scenes. First, we adapt a new input space from raw RGB image space by exploiting underwater light attenuation prior, and then devise a least-squared formulation for coarse pixel-wise depth prediction. Subsequently, we extend this into a domain projection loss that guides the end-to-end learning of UDepth on over 9K RGB-D training samples. UDepth is designed with a computationally light MobileNetV2 backbone and a Transformer-based optimizer for ensuring fast inference rates on embedded systems. By domain-aware design choices and through comprehensive experimental analyses, we demonstrate that it is possible to achieve state-of-the-art depth estimation performance while ensuring a small computational footprint. Specifically, with 70 % −80 % less network parameters than existing benchmarks, UDepth achieves comparable and often better depth estimation performance. While the full model offers over 66 FPS (13 FPS) inference rates on a single GPU (CPU core), our domain projection for coarse depth prediction runs at 51.5 FPS rates on single-board Jetson TX2s. The inference pipelines are available at https://github.com/uf-robopi/UDepth.
1 December 2021 in Case Studies in Construction Materials
Production of plastic has risen significantly in the last 60 years worldwide, and around 10% of this plastic is turned into solid waste. This plastic becomes an environmental hazard in the absence of appropriate recovery methods. Incorporating waste plastic in concrete is an alternate way to recycle this waste plastic. Polypropylene (PP) and polyethylene terephthalate (PET) are two of the most widely used plastic scarcely recovered and recycled. Therefore, in the present study, both PP and PET are used as a partial replacement (10%, 20%, and 30% by volume) of coarse aggregate. These concretes are compared for three different water‐cement ratios (0.42, 0.48, and 0.57). Properties such as workability, density, compressive, and tensile strengths are compared. Concrete with PP aggregate has shown as much as 39% higher compressive strength and 9% lower density than brick aggregate concrete. On the other hand, concrete with PET aggregate has displayed as much as 53% reduction in compressive strength compared to the reference concrete. Furthermore, it demonstrates higher workability and lower density. Equations are proposed to describe relationships between compressive and splitting tensile strengths of concrete, including the effect of plastic aggregate percentages. Cost analysis of the concrete reveals that incorporating waste plastic in concrete is more expensive than the regular brick aggregate concrete. Still, up to 10%, PP can be used in concrete, considering the gain in compressive strength.