日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

経路計画arXiv:0704.3268

単層興奮性CNNモデルによる2次元経路解

2D Path Solutions from a Single Layer Excitable CNN Model

シェア:XThreadsFacebookLINEはてブBluesky

興奮性・双安定性を持つ単層の細胞非線形ネットワーク(CNN)を用い、障害物分布を直接マッピングして2次元空間の経路解を計算するアルゴリズムを提案し、回路レベルのパラメータで数値実験した。

著者: Koray Karahaliloglu

分類: cs.RO, cs.NE

原文アブストラクト

An easily implementable path solution algorithm for 2D spatial problems, based on excitable/programmable characteristics of a specific cellular nonlinear network (CNN) model is presented and numerically investigated. The network is a single layer bioinspired model which was also implemented in CMOS technology. It exhibits excitable characteristics with regionally bistable cells. The related response realizes propagations of trigger autowaves, where the excitable mode can be globally preset and reset. It is shown that, obstacle distributions in 2D space can also be directly mapped onto the coupled cell array in the network. Combining these two features, the network model can serve as the main block in a 2D path computing processor. The related algorithm and configurations are numerically experimented with circuit level parameters and performance estimations are also presented. The simplicity of the model also allows alternative technology and device level implementation, which may become critical in autonomous processor design of related micro or nanoscale robotic applications.

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. 3Google DeepMind、ロボット向けAIを発表新製品・発表
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5GXO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

PR・アフィリエイト広告

OZ GAMING(GPU搭載BTOゲーミングPC)

OZ GAMING(GPU搭載BTOゲーミングPC)の公式広告バナーOZ GAMINGで完成品PCを見る ↗
PR本紙発行元 EmplifAI