日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

自己位置推定arXiv:2405.02290

車輪オドメトリによる自律走行車椅子の自己位置推定

Wheel Odometry-Based Localization for Autonomous Wheelchair

シェア:XThreadsFacebookLINEはてブBluesky

GPSに頼らず、車輪の回転情報から車椅子の位置と向きを推定する手法を評価した研究。

著者: P Paryanto, Rakha Rahmadani Pratama, Roni Permana Saputra

分類: cs.RO

原文アブストラクト

Localization is a fundamental requirement for an autonomous vehicle system. One of the most often used systems for autonomous vehicle localization is the global positioning system (GPS). Nevertheless, the functionality of GPS is strongly dependent on the availability of satellites, making it unreliable in some situations. As a result, autonomous vehicles must possess autonomous self-localization capabilities to ensure their independent operation. Odometry techniques are employed to achieve vehicle localization by predicting the vehicle position and orientation based on sensor measurements of the vehicle motion. One of the approaches employed in odometry is known as wheel odometry. Wheel odometry has a lower degree of reliance on the surrounding environment than visual odometry and laser odometry. This study aims to evaluate the performance of wheel odometry implementation for an autonomous wheelchair in the context of the localization process. The differential drive kinematic model is employed to determine the predicted pose of a wheelchair. This prediction is derived from the measurement of the linear and angular velocity of the wheelchair. Several experiments have been conducted to evaluate the performance of wheel odometry-based localization. Prior to experimenting, calibration procedures have also been performed to ensure accurate measurements of the sensor.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  3. 3GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5UC BerkeleyとNVIDIAの研究者、AIエージェントでロボット制御コードを仮想環境で改善研究・技術

PR・アフィリエイト広告

FRONTIER(GPU搭載BTOパソコン)

FRONTIER(GPU搭載BTOパソコン)の公式広告バナーFRONTIERでGPU搭載PCを見る ↗
PR本紙発行元 EmplifAI