Skip to content
Trending eventDeveloping

SketchMate: Agentic Geometric Reasoning for Zero-Shot Precision Insertion

1 reports1 reporting sourcesUpdated 6 hours ago

Overview

Event synthesis

This research introduces SketchMate, a system that uses vision-language models to create geometric sketches of insertion tasks. By decoupling part description from placement and using sketch memory with point-cloud registration, the method achieves sub-millimetre precision in simulated environments. On 69 episodes, it successfully inserted 88% of unseen parts at the correct pose, significantly outperforming existing zero-shot agentic manipulation methods which failed on the same task.

Generated from attributed reports · Updated 2 hours ago

Event evidence and corrections

0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.

Environment: simulation
Supporting report

“asping the part, the robot inspects both mating faces at close range and corrects the sketch by shape matching. On 69 simulated insertion episodes, our method inserts 88% of the parts at the assembled pose in a strict kinematic test, while ReKep and MOKA with the same VLM insert none. On new placements of eight tasks, reusing one stored run inserts as many parts as calling the agent in each episode, and it produces t”

Exact source · revision 1

Source owner not reported

Reported quantity · sketch generation time: 3.4 other · Basis not reported
Supporting report

“he sketch in 3.4 s instead of about 90 s.”

Exact source · revision 1

Source owner not reported

Reported quantity · baseline sketch generation time: 90 other · Basis not reported
Supporting report

“he sketch in 3.4 s instead of about 90 s.”

Exact source · revision 1

Source owner not reported

Reported quantity · simulated episodes: 69 other · reported trials
Supporting report

“asping the part, the robot inspects both mating faces at close range and corrects the sketch by shape matching. On 69 simulated insertion episodes, our method inserts 88% of the parts at the assembled pose in a strict kinematic test, while ReKep and MOKA with the same VLM insert none. On new placements of eight tasks, reusing one stored run inserts as many parts as calling the agent in each episode, and it produces t”

Exact source · revision 1

Source owner not reported

Reported quantity · number of tasks: 8 other · Basis not reported
Supporting report

“asping the part, the robot inspects both mating faces at close range and corrects the sketch by shape matching. On 69 simulated insertion episodes, our method inserts 88% of the parts at the assembled pose in a strict kinematic test, while ReKep and MOKA with the same VLM insert none. On new placements of eight tasks, reusing one stored run inserts as many parts as calling the agent in each episode, and it produces t”

Exact source · revision 1

Source owner not reported

Reported quantity · robots: 1 other · Basis not reported
Supporting report

“asping the part, the robot inspects both mating faces at close range and corrects the sketch by shape matching. On 69 simulated insertion episodes, our method inserts 88% of the parts at the assembled pose in a strict kinematic test, while ReKep and MOKA with the same VLM insert none. On new placements of eight tasks, reusing one stored run inserts as many parts as calling the agent in each episode, and it produces t”

Exact source · revision 1

Source owner not reported

Report timeline

Follow attributed reports and material updates.

11/2
  1. arXiv Robotics — research abstractsSignal
    SketchMate: Agentic Geometric Reasoning for Zero-Shot Precision Insertion

    This research introduces SketchMate, a system that uses vision-language models to create geometric sketches of insertion tasks. By decoupling part description from placement and using sketch memory with point-cloud registration, the method achieves sub-millimetre precision in simulated environments. On 69 episodes, it successfully inserted 88% of unseen parts at the correct pose, significantly outperforming existing zero-shot agentic manipulation methods which failed on the same task.

Event coverage history

There is not enough continuous observation data to show a trend.

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).