A high-precision object pose estimation and fine manipulation method based on visual-tactile fusion

By defining master and slave objects and combining particle filtering technology and tactile information, the posture changes of slave objects are estimated in real time, which solves the problem of low accuracy of visual posture estimation in dynamic environments and achieves high-precision object posture estimation and fine operation.

CN119458332BActive Publication Date: 2025-09-26BEIJING INST OF TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202411637620.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-11-15
Publication Date
2025-09-26
Estimated Expiration
2044-11-15

AI Technical Summary

Technical Problem

Existing vision-based object pose estimation methods have low accuracy in dynamic environments, and traditional vision-tactile fusion methods fail to effectively deal with pose changes caused by object sliding or movement.

Method used

By defining a master object and a slave object, the posture changes of the master object are tracked in real time. Particle filtering technology is combined with tactile information to estimate the posture changes of the slave object relative to the robot's end effector, including translational and rotational slippage. Tactile sensors are used to detect sliding signals, and collision simulation and particle filtering algorithms are combined to perform accurate posture estimation.

Benefits of technology

It achieves high-precision pose estimation in dynamic environments, with a translation error of 1-2 mm and a rotation error of approximately 2 degrees. It is suitable for a variety of dynamic operation scenarios and improves the robot's ability to respond in dynamic interactive environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention provides a high-precision object pose estimation and fine manipulation method based on visual-tactile fusion, which obtains visual and tactile information through the physical interaction between the robot and the environment, thereby completing accurate pose estimation in dynamic scenes; the present invention constructs a hand object pose correction algorithm based on particle filtering, which can estimate the pose change when the held object slides and the contacted object moves. This method is suitable for most high-precision operation scenarios, including some challenging scenarios with dynamic attribute changes, and significantly improves the robot's ability to cope with dynamic interactive environments; the present invention performs high-precision operation tasks on 13 objects with different geometric shapes, including the FMB benchmark. The results show that our method achieves accurate pose estimation with a translation error of 1-2 mm and a rotation error of about 2 degrees. At the same time, it can cope with dynamic situations such as object sliding.
Need to check novelty before this filing date? Find Prior Art