Event
RobotAgentBench: An Open-World Robot Autonomy Benchmark Released
1 reports · 1 sources
Lingjun Mao and other authors released RobotAgentBench (RAB), a robot agent evaluation benchmark for real open-world household environments. Through an agent-assisted, human-in-the-loop real-to-simulation pipeline, the benchmark constructs 50 household layouts, including 90 atomic tasks, 18 short-horizon tasks, 8 long-horizon tasks, and 4 open-ended scenarios. The authors state that RAB can systematically compare standalone VLA/WAM models with various robot agent architectures under a unified planning-execution-verification framework.
Report timeline
- RobotAgentBench released: a benchmark for open-world robot autonomy
The authors propose RobotAgentBench (RAB), a benchmark for evaluating robot agents in realistic open-world household environments. Using an agent-assisted, human-in-the-loop real-to-sim pipeline, it constructs 50 household layouts and includes 90 atomic tasks, 18 short-horizon tasks, 8 long-horizon tasks, and 4 open-ended scenarios.
arXiv cs.RO ·