Search Results

In this paper, we introduce the Semantic Environment Atlas (SEA), a novel mapping approach designed to enhance visual navigation capabilities of embodied agents. The SEA utilizes semantic graph maps that intricately delineate the relationships ...

• Enhanced Navigation Capabilities: The SEA enhances navigation using semantic graph maps, improving localization and tasks.
• Robust Against Sensor Noise: SEA achieves robust navigation by leveraging semantic knowledge, even with sensor ...

Reinforcement learning from human feedback (RLHF) alleviates the problem of designing a task-specific reward function in reinforcement learning by learning it from human preference. However, existing RLHF models are considered inefficient as they produce ...

In safety-critical robotic tasks, potential failures must be reduced, and multiple constraints must be met, such as avoiding collisions, limiting energy consumption, and maintaining balance. Thus, applying safe reinforcement learning (RL) in such robotic ...

Addressing decision-making problems using sequence modeling to predict future trajectories shows promising results in recent years. In this paper, we take a step further to leverage the sequence predictive method in wider areas such as long-term planning,...

• We propose a coarse depth estimation method for compound eye images based on deep learning.
• We propose a network suitable for compound eye structure based on Vision Transformer.
• Performance of coarse depth estimation improves ...

A compound eye camera is a hemispherical camera made by mimicking the structure of an insect’s eye. In general, a compound eye camera is composed of a set of single eye cameras. The compound eye camera has various advantages due to its unique ...

This paper emphasizes the importance of a robot's ability to refer to its task history, especially when it exe-cutes a series of pick-and-place manipulations by following language instructions given one by one. The advantage of referring to the ...

Imitation learning is a methodology, which trains an agent using demonstrations from skilled experts without external rewards. However, for a complex task with a long horizon, it is challenging to obtain data that exactly match the desired task. ...

This paper tackles the image-goal navigation problem, in which a robot needs to find a goal pose based on the target image. The proposed algorithm estimates the geometric information between the target pose and the current pose of the robot. Using the ...

In this paper, we tackle the problem of efficient data-driven control of a system, of which the observations are given by raw images. By learning latent representations of image observations and their dynamics model, the image-based control problems can ...

In this paper, we tackle the problem of Image-Goal Navigation, where an agent is asked to find the viewpoint of a given goal image in an unseen environment. Past methods rely on depth cameras, require panoramic observations and a pre-built graph network, ...

In this paper, we address a planar non-prehensile rearrangement task. As a problem of pushing objects to desired target points, we model the problem as Multi-objective Markov decision processes (MOMDPs) to efficiently solve it and propose a method of ...

The vision-based 3D reconstruction methods have various advantages and can be used in various applications such as navigation. Although various vision-based methods are being studied, it is difficult to reconstruct many parts at once with a general camera ...

Visual place localization (VPL) is a problem finding the closest database image from a query image. Since the outdoor images can be recognized from GPS sensors, VPL in an indoor scene is a difficult problem. Also, Image changes indoors are more severe ...

We propose a novel autonomous driving frame-work that leverages graph-based features of roads, such as road positions and connections. The proposed method is divided into two parts: a low-level controller which follows the trajectory calculated by a graph-...

Much progress has been made for non-rigid structure from motion (NRSfM) during the last two decades, which made it possible to provide reasonable solutions for synthetically-created benchmark data. In order to utilize these NRSfM techniques in more ...

In this paper, we consider stochastic multi-armed bandits (MABs) with heavy-tailed rewards, whose p-th moment is bounded by a constant ν_p for 1 < p ≤ 2. First, we propose a novel robust estimator which does not require ν_p as prior information, while ...

In this paper, we propose a novel contextual bandit algorithm that employs a neural network as a reward estimator and utilizes Shannon entropy regularization to encourage exploration, which is called Shannon entropy regularized neural contextual bandits (...

In this paper, we consider autonomous driving of a vehicle using imitation learning. Generative adversarial imitation learning (GAIL) is a widely used algorithm for imitation learning. This algorithm leverages positive demonstrations to imitate the ...

This paper proposes a multiple stakeholder perspective model (MSPM) which predicts the future pedestrian trajectory observed from vehicle&#x2019;s point of view. For the vehicle-pedestrian interaction, the estimation of the pedestrian&#x2019;s intention ...

The goal of the optimal viewpoint path estimation is to generate a path to the optimal viewpoint location where the robot can best see the Point of Interest (POI). There are several learning-based methods to find an optimal viewpoint, but these methods ...