RoboAtlas: Contextual Active SLAM
RoboAtlas: Contextual Active SLAM
RoboAtlas is a contextual Active SLAM framework that adaptively balances geometric exploration and semantic reasoning using a scalable 3D semantic mapping system. Evaluated in simulation and on a Unitree Go2 robot in large-scale real-world environments, it achieves a 100% task success rate and state-of-the-art performance on the GOAT-Bench 'Val Unseen' benchmark, with success rates of 90.6% using GPT-4o and 88.8% using Qwen2.5-VL-7B.
RoboAtlas introduces a contextual Active SLAM framework that combines geometric exploration with semantic reasoning using a 3D semantic mapping system. The system demonstrates strong performance in both simulation and real-world environments, achieving high task success rates and outperforming prior baselines, particularly when using smaller, more efficient models.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
mapped semantic instances
30,000
Reported success rate
90.6%
semantic instances mapped
30,000
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
to semantically guided navigation as scene understanding improves. We evaluate the system in simulation and on a Unitree Go2 robot in large-scale real-world environments exceeding 1800 m2 with approx. 30k mapped semantic instances, achieving a 100% task success rate. On the GOAT-Bench "Val Unseen" benchmark, RoboAtlas achieves state-of-the-art performance with highest reported success rate (SR) of 90.6%, using GPT-4o
Open source S3
, improving over the strongest prior baseline by 17.8 percentage points in SR. Using the much smaller Qwen2.5-VL-7B model, it still achieves 88.8% SR, outperforming all baselines using GPT-4o in SR, and revealing the importance of the information gained by our semantic mapping framework over simply replacing the underlying foundation model. The results demonstrate that grounding foundation models with large-scale 3D
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org