DexArena benchmark tests robot skill for long tasks with hands
DexArena: A Unified Benchmark for Long-Horizon Dexterous Manipulation
Robotics
Summary
Robots have trouble doing tricky tasks that need many steps using both hands, like you might in your kitchen or at work. The researchers created DexArena, a set of tests to see how well robot software can learn and handle these complicated two-handed tasks. It includes 10 big tasks broken into smaller parts, letting developers check if robots are making progress or finishing entire jobs. The benchmark uses realistic simulations and real human-controlled examples to make sure tests are fair and results repeatable.
What this means in practice
- •For robotics engineers: Evaluate and improve robot control software for complex two-handed tasks in simulated household and industrial settings.
- •For automation system designers: Test how automation algorithms handle chaining multiple precise manipulation skills across varied environments.
Authors
Haonan Ping, Jianshu Hu, Yimin Wang, Yi Zhu, Pai Liang, Qiyan Jin, Qirui Bao, Ying Qian, Xinyang Ding, Hongfei Qiao, Tianle Xue, Yucheng Bai, Soo Wei Jie, Shuguang Li, Yutong Ban
Abstract
Long-horizon dexterous manipulation remains a fundamental challenge for robot learning, requiring the composition of multiple skills, precise bimanual coordination, generalizable visual understanding, and robust skill transitions. Yet existing manipulation benchmarks largely emphasize isolated skills or support only a limited range of end-effectors, making these capabilities difficult to evaluate systematically. We introduce DexArena, a benchmark designed to evaluate learning and generalization in long-horizon bimanual dexterous manipulation. DexArena features 10 long-horizon tasks and 30 stage-level tasks spanning diverse household and industrial scenarios. Each long-horizon task is composed of explicitly defined stages with physically grounded success conditions, enabling reliable evaluation of both intermediate progress and end-to-end task completion. A unified benchmark interface and verified human-teleoperated demonstrations enable testing different policy learning algorithms, with standardized simulation configurations, scene randomization, trajectory replay, and task-level evaluation. Through extensive baseline experiments, we identify substantial challenges in precise manipulation, generalization to scene variation, and skill composition. DexArena establishes a challenging and reproducible benchmark for developing robot learning methods that move toward generalizable long-horizon dexterous behavior.