Eri SAWADA

JSK Robotics Lab, The University of Tokyo

[Eri SAWADA @erisawada3030]

卒論の成果が #IROS2026 に採択されました！

ソフト要素を組み込んだ変形飛行ロボットで、空中マニピュレーションをします！

[Eri SAWADA @erisawada3030]

🎉 Our new paper, “A Morphing Aerial Robot With Thruster-Integrated Flexible Continuum Links for Shape Adaptive Aerial Manipulation,” has been accepted to #IROS2026!

Our novel aerial robot achieves versatile aerial manipulation via shape adaptability & expanded wrench space!

[Eri SAWADA @erisawada3030]

🔑𝗞𝗲𝘆 𝗛𝗶𝗴𝗵𝗹𝗶𝗴𝗵𝘁𝘀:
• Hardware: Prevents unintended bending while preserving horizontal flexibility.
• Control: Dual strategies for stable flight and in-air morphing.
• Experimental Validation: Proven performance in challenging tasks!

[Eri SAWADA @erisawada3030]

🎉 Our new paper, “A Morphing Aerial Robot With Thruster-Integrated Flexible Continuum Links for Shape Adaptive Aerial Manipulation,” has been accepted to #IROS2026!

Our novel aerial robot achieves versatile aerial manipulation via shape adaptability & expanded wrench space!

[Eri SAWADA @erisawada3030]

📜Paper: https://t.co/DcrJQ80cQo
📹Video: https://t.co/PDxOsjgTBg

[OptunaAutoML @OptunaAutoML]

Eri Sawada explains parallel Bayesian optimization based on the q-Batch acquisition function using the Monte Carlo method, introduced in Optuna v5. Read the article here: https://t.co/ZQURT28SsI

[PFN Tech @preferred_jp]

【ブログ】Optuna開発のアルバイトとしてPFNで勤務された澤田恵里さんによる寄稿記事です。本記事では、Optuna v5に導入されたモンテカルロ法を用いたq-batch獲得関数に基づく並列ベイズ最適化について解説しています。 @OptunaAutoML https://t.co/7UIty3sW1D

[Eri SAWADA @erisawada3030]

@OptunaAutoML OptunaのGPSamplerにq-バッチ獲得関数を追加しました！日本語版のブログはこちらです。https://t.co/dTGQNRUlkj

[Eri SAWADA @erisawada3030]

In @OptunaAutoML v5.0, I introduced q-batch acquisition functions to GPSampler to enhance parallel optimization!
I wrote a blog post explaining the motivation behind the method, algorithm details, and benchmark performance.
Read more here: https://t.co/dIqBVTPCVI

[Eri SAWADA @erisawada3030]

In @OptunaAutoML v5.0, I introduced q-batch acquisition functions to GPSampler to enhance parallel optimization!
I wrote a blog post explaining the motivation behind the method, algorithm details, and benchmark performance.
Read more here: https://t.co/dIqBVTPCVI

[Eri SAWADA @erisawada3030]

#robomech2026 明日7月1日に発表します！

2P1-C11
「柔軟連続体リンク構造を搭載した変形飛行ロボットの制御と柔軟性のタスク応用」

柔軟リンクを有する機体で安定飛行を実現し、様々な空中マニピュレーションを行いました！よろしくお願いします！！

[OptunaAutoML @OptunaAutoML]

Eri Sawada introduces the constant-liar and Kriging-believer approaches to Optuna's GPSampler. These methods enable efficient parallelization for single-objective, multi-objective, and constrained optimization. Learn more in our article: https://t.co/thPAVLfwra