From Task Execution to Autonomous User Needs: RobotAgentBench Evaluates Open-World Robot Autonomy
当机器人走出预设任务：RobotAgentBench全面评测机器人的开放世界能力
A research paper introduces RobotAgentBench, a benchmark that evaluates the open-world autonomy of household robots. It assesses five key dimensions of robot capability, including operation, task discovery, context reasoning, failure recovery, and robustness. The study compares different agent architectures, showing that hybrid execution strategies and recursive self-improvement (RSI) significantly enhance performance. Results indicate that while independent VLA/WAM models struggle with open-ended tasks, integrating code-based planning and learning strategies improves task completion and adaptability.
RobotAgentBench introduces a comprehensive benchmark for evaluating open-world autonomy in household robots, highlighting the gap between task execution and autonomous reasoning. The study compares various agent architectures and demonstrates the benefits of integrating planning, execution, and validation frameworks.
Source: Chinese robotics — Research news discovery · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
family layouts
50
View original evidence
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11atomic tasks
90
View original evidence
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11short tasks
18
View original evidence
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11long tasks
8
View original evidence
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11open scenarios
4
View original evidence
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11family layouts
50
rooms
50
furniture configurations
50
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
它以真实家庭为起点，构建 50 种家庭布局、90 项原子任务、18 项短程任务、8 项长程任务和 4 类开放式场景，系统比较独立 VLA/WAM 策略与多种机器人智能体架构。
Open source S11
通过这一流程，RobotAgentBench 构建了 50 种家庭布局，覆盖不同的房间结构、装修、墙面、地板与家具配置。
Open source S21
Source：Chinese robotics — Research news discovery · finance.sina.com.cn