Search

Navigation in primates is generally supported by cognitive maps. Such a map endows an animal with navigational planning capabilities. Numerous methods have been proposed to mimic these natural navigation capabilities in artificial systems. Based on self-navigation and learning strategies in animals, we propose in this work a place recognition strategy for brain-inspired navigation. First, a place recognition algorithm structure based on convolutional neural networks (CNNs) is introduced, which can be applied in the field of intelligent navigation. Second, sufficient images are captured at each landmark and then stored as a reference image library. Simple linear iterative clustering (SLIC) is used to segment each image into superpixels with multi-scale viewpoint-invariant landmarks. Third, highly representative appearance-independent features are extracted from these landmarks through CNNs. In addition, spatial pyramid pooling (SPP) layers are introduced to generate a fixed-length CNN representation, regardless of the image size. This representation boosts the quality of the extracted landmark features. The proposed SLIC-SPP-CNN place recognition algorithm is evaluated on one collected dataset and two public datasets with viewpoint and appearance variations. [ABSTRACT FROM AUTHOR]

The traditional processing model of the temperature error for a gyroscope is serial, meaning that de-noising and temperature drift compensation are implemented in a two-step procedure. Hence, the result of the latter depends on the performance of the former; in particular, negative de-noising produces a negative compensation result. To reduce this dependence, we propose a parallel processing algorithm of the temperature error based on variational mode decomposition (VMD) and an augmented nonlinear differentiator (AND). An application to a micro-electro-mechanical system gyroscope is described to demonstrate the effectiveness and applicability of the proposed algorithm. Its major advantages are (i) a combination of VMD, extreme learning machines, and AND is proposed, and an adaptive accelerometer factor determination method for AND is given based on the VMD, both of which improve the effectiveness of the de-noising process; (ii) temperature drift and noise in the temperature error can be extracted and processed synchronously, thereby reducing the dependency of drift compensation on the de-noising result and making the temperature error process more efficient. [ABSTRACT FROM AUTHOR]

A noise reduction algorithm based on an improved empirical mode decomposition (EMD) and forward linear prediction (FLP) is proposed for the fiber optic gyroscope (FOG). Referred to as the EMD-FLP algorithm, it was developed to decompose the FOG outputs into a number of intrinsic mode functions (IMFs) after which mode manipulations are performed to select noise-only IMFs, mixed IMFs, and residual IMFs. The FLP algorithm is then employed to process the mixed IMFs, from which the refined IMFs components are reconstructed to produce the final de-noising results. This hybrid approach is applied to, and verified using, both simulated signals and experimental FOG outputs. The results from the applications show that the method eliminates noise more effectively than the conventional EMD or FLP methods and decreases the standard deviations of the FOG outputs after de-noising from 0.17 to 0.026 under sweep frequency vibration and from 0.22 to 0.024 under fixed frequency vibration.