Search

Recent visual odometry (VO) methods incorporating geometric algorithm into deep-learning architecture have shown outstanding performance on the challenging monocular VO task. Despite encouraging results are shown, previous methods ignore the requirement of generalization capability under noisy environment and various scenes. To address this challenging issue, this work first proposes a novel optical flow network (PANet). Compared with previous methods that predict optical flow as a direct regression task, our PANet computes optical flow by predicting it into the discrete position space with optical flow probability volume, and then converting it to optical flow. Next, we improve the bundle adjustment module to fit the self-supervised training pipeline by introducing multiple sampling, ego-motion initialization, dynamic damping factor adjustment, and Jacobi matrix weighting. In addition, a novel normalized photometric loss function is advanced to improve the depth estimation accuracy. The experiments show that the proposed system not only achieves comparable performance with other state-of-the-art self-supervised learning-based methods on the KITTI dataset, but also significantly improves the generalization capability compared with geometry-based, learning-based and hybrid VO systems on the noisy KITTI and the challenging outdoor (KAIST) scenes., Comment: 35 pages, 6 figures

Sodium-ion batteries show promising candidate for next-generation battery due to the low cost and abundance of sodium sources. Here, we have successfully fabricated bimetallic cobalt/zinc sulfides (Co4S3/Co9S8/ZnS) microflowers via a facile solvothermal method, and further evaluated their sodium storage performance. Due to bimetallic compounds and pseudocapacitive effect, the composites exhibit excellent electrochemical performance including high specific capacity, high rate capability and long-term cyclic stability.

One mild multiple emulsions was used for the fabrication of hollow polystyrene (PS) spheres. Polystyrene was dissolved in a volatile organic solvent to form the O phase, then an aqueous phase containing a surfactant was transferred into above phase to form an oil-in-water (W in /O) emulsion, followed by addition of W in /O emulsion in the external W out phase, and solvent evaporation. Final product with diameter of 150 μm, surface roughness of 30 nm, and sphericity of 98.0% was obtained. The possible influences on the formation of PS hollow sphere were also discussed.