Convolutional Neural Networks (CNN) has been widely applied in the realm of computer vision. However, given the fact that CNN models are translation invariant, they are not aware of the coordinate information of each pixel. Thus the generalization ability of CNN will be limited since the coordinate information is crucial for a model to learn affine transformations which directly operate on the coordinate of each pixel. In this project, we proposed a simple approach to incorporate the coordinate information to the CNN model through coordinate embedding. Our approach does not change the downstream model architecture and can be easily applied to the pre-trained models for the task like object detection. Our experiments on the German Traffic Sign Detection Benchmark show that our approach not only significantly improve the model performance but also have better robustness with respect to the affine transformation.
This paper presents the modelling and experimental evaluation of a semi-active vehicle suspension installed with a self-powered MR damper which is able to perform variable stiffness. Its variable stiffness feature as well as the self-powering capability was evaluated and verified using a hydraulic Instron test system. The testing results show that the stiffness of the damper is dependent on the current which can be generated by the self-powering component. A mathematic model was established to describe the dynamic properties of the MR damper and its power-generating capability. Finally, the self-powered MR suspension was installed on a quarter car test rig for its vibration isolation evaluation. A controller based on the short-time Fourier transform (STFT) was developed for the stiffness control. The evaluation result illustrates that the proposed MR damper can reduce the acceleration and displacement of the sprung mass by 16.8% and 21.4% respectively, compared with the passive system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.