視覚的把持を超えて:検出から実行までの複雑な把持のベンチマーキング
Beyond Visual Grasping: Benchmarking Complex Grasping from Detection to Execution
複雑な実世界の把持タスクを評価するための新しいベンチマークGCA-Benchを提案し、従来の視覚ベースの把持検出を超えて、シーン推論と意味的制約を含むタスクでの大規模基盤モデルの性能を評価した。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Hanyi Zhang, Khang Nguyen, Charith Munasinghe, Basu Hela, Tianyu Li, Zihong Luo, Hoan Nguyen, Hans Wernher van de Venn, Yalin Zheng, Ravi Prakash, Tung D. Ta, Anh Nguyen, Baoru Huang
分類: cs.RO, cs.AI
原文アブストラクト
Robust robotic grasping remains a fundamental challenge for complex real-world applications. Recent advances in large-scale models demonstrate promising capabilities for reasoning in robotic tasks. However, existing benchmarks for grasping primarily focus on isolated, visual-based grasp pose detection, failing to capture the complexity of grasping tasks that require multi-step reasoning and semantic understanding during execution. To address this gap, we propose GCA-Bench, a benchmark featuring challenging \textit{grasping with complex action} scenarios that involve both scene-level reasoning and semantic constraints. GCA-Bench enables the evaluation of recent large foundation models under the same settings. To demonstrate the effectiveness of our new benchmark, we implement a diverse set of baselines, ranging from traditional grasp detection pipelines to end-to-end learning methods. Empirical studies achieve success rates below 70\% on complex grasping scenarios, underscoring critical limitations. In addition, we propose new evaluation metrics, analyze critical failure models, and provide insights to guide the development of more robust and generalizable grasping strategies.
