日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

週刊ニュースレター購読
歩行arXiv:2603.05902

小型ロボットの傾斜地ホッピング制御の改善:スプリングマスモデルを用いたアプローチ

Improved hopping control on slopes for small robots using spring mass modeling

シェア:XThreadsFacebookLINEはてブBluesky

傾斜地でのホッピングロボットの不安定化をスプリングマスモデルで解析し、着地角度の調整と離陸前の補正トルクという簡単な対策で安定性を向上させる手法を提案した。

著者: Heston Roberts, Pronoy Sarker, Sm Ashikul Islam, Min Gyu Kim

分類: cs.RO, cs.SY, eess.SY

原文アブストラクト

Hopping robots often lose balance on slopes because the tilted ground creates unwanted rotation at landing. This work analyzes that effect using a simple spring mass model and identifies how slope induced impulses destabilize the robot. To address this, we introduce two straightforward fixes, adjusting the bodys touchdown angle based on the slope and applying a small corrective torque before takeoff. Together, these steps effectively cancel the unwanted rotation caused by inclined terrain, allowing the robot to land smoothly and maintain stable hopping even on steep slopes. Moreover, the proposed method remains simple enough to implement on low cost robotic platforms without requiring complex sensing or computation. By combining this analytical model with minimal control actions, this approach provides a practical path toward reliable hopping on uneven terrain. The results from simulation confirm that even small slope aware adjustments can dramatically improve landing stability, making the technique suitable for future autonomous field robots that must navigate natural environments such as hills, rubble, and irregular outdoor landscapes.

関連論文