In prior research, a method was introduced for traversability perception in autonomous mobile robots, utilizing a Growing Neural Gas with Different Topologies (GNG-DT). The traversability refers to the points at which an autonomous mobile robot can traverse. GNG-DT traditionally relied on pre-set parameters like slope angle and surface roughness to estimate traversability. To address this limitation, an estimation technique leveraging supervised learning is proposed. Using the proposed method, it is shown that the robot performance-based traversability can be estimated using supervised learning. In addition, the robot can avoid obstacles using self-learned traversability. The current study advances this work by introducing a self-learning approach for estimating traversability, enabling self-supervised learning in perceptual systems.

The problem of image change detection via every-day indoor robot navigation is explored from a novel perspective of the self-attention technique. Detecting semantically non-distinctive and visually small changes remains a key challenge in the robotics community. Intuitively, these small non-distinctive changes may be better handled by the recent paradigm of the attention mechanism, which is the basic idea of this work. However, existing self-attention models require significant retraining cost per domain, so it is not directly applicable to robotics applications. We propose a new self-attention technique with an ability of unsupervised on-the-fly domain adaptation, which introduces an attention mask into the intermediate layer of an image change detection model, without modifying the input and output layers of the model. Experiments, in which an indoor robot aims to detect visually small changes in everyday navigation, demonstrate that our attention technique significantly boosts the state-of-the-art image change detection model. Our datset is available at https://github.com/KojiTakeda00/Small_object_change_detection

This paper describes a method to detect localization failure for mobile robots. In spite of advances in localization methods, mobile robots still may fail in estimating its pose due to degeneration of observations, or unexpected objects around them obstructing observations. We propose a method for mobile robots to monitor these kinds of failures and then take actions for restoring correct localization status.

Effective GPU implementations of an inverse iteration algorithm with reorthogonalization are proposed for computing eigenvectors of symmetric tridiagonal matrices. The key to effectively accelerating the inverse iteration algorithm in GPU computing is the adoption of reorthogonalization code optimal for the GPU. The CGS2 algorithm and the compact WY orthogonalization algorithm, which can be implemented using level 2 BLAS routines, are implemented using CUBLAS. The size of the data transferred between the CPU and GPU is also optimally reduced. The proposed code of the inverse iteration algorithm using the CGS2 algorithm is shown to map well to a GPU and to achieve high performance through numerical experiments on a CPU-GPU heterogeneous computer.