日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

乗り物酔い評価arXiv:2412.14982

コンパクトテストコースで実路運転を再現する効率的な乗り物酔い評価手法

Efficient Motion Sickness Assessment: Recreation of On-Road Driving on a Compact Test Track

シェア:XThreadsFacebookLINEはてブBluesky

モデル予測制御を用いて実路走行時の加速度を小型テストコース上で再現し、乗り物酔いの発生が実路とテストコースで同等であることを47名の被験者実験で示した。

著者: Huseyin Harmankaya, Adrian Brietzke, Rebecca Pham-Xuan, Barys Shyrokau, Riender Happee, Georgios Papaioannou

分類: cs.RO, cs.ET, cs.HC

原文アブストラクト

The ability to engage in other activities during the ride is considered by consumers as one of the key reasons for the adoption of automated vehicles. However, engagement in non-driving activities will provoke occupants' motion sickness, deteriorating their overall comfort and thereby risking acceptance of automated driving. Therefore, it is critical to extend our understanding of motion sickness and unravel the modulating factors that affect it through experiments with participants. Currently, most experiments are conducted on public roads (realistic but not reproducible) or test tracks (feasible with prototype automated vehicles). This research study develops a method to design an optimal path and speed reference to efficiently replicate on-road motion sickness exposure on a small test track. The method uses model predictive control to replicate the longitudinal and lateral accelerations collected from on-road drives on a test track of 70 m by 175 m. A within-subject experiment (47 participants) was conducted comparing the occupants' motion sickness occurrence in test-track and on-road conditions, with the conditions being cross-randomized. The results illustrate no difference and no effect of the condition on the occurrence of the average motion sickness across the participants. Meanwhile, there is an overall correspondence of individual sickness levels between on-road and test-track. This paves the path for the employment of our method for a simpler, safer and more replicable assessment of motion sickness.

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2GXO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
  3. 3newmo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
  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