Event
AffordSim: An Affordance-Aware Manipulation Data Generator and Benchmark
1 reports · 1 sources
Mingyang Li and co-authors released AffordSim, a data generator and benchmark that connects open-vocabulary 3D affordance prediction to simulated trajectory generation. Its pipeline is: synthesize scenes according to natural-language task descriptions, generate affordance queries, sample region-constrained grasps, and then use motion planning to filter executable candidates. The authors state that AffordSim contains 50 tasks, covering multiple manipulation skills, five robot embodiments, and more than 500 rigid and articulated objects. Vision-language-action policies trained on AffordSim data can transfer zero-shot to a real Franka FR3, with an average success rate of 24%.
Report timeline
- AffordSim: A Scalable Data Generator and Benchmark for Affordance-Aware Robotic Manipulation
AffordSim is a scalable data generator and benchmark that integrates open-vocabulary 3D affordance prediction into simulation trajectory generation. Given a natural-language task description, it synthesizes scenes, generates affordance queries, and samples region-constrained grasps, then uses motion planning to filter executable candidates.
arXiv cs.RO ·