Massachusetts Institute of Technology

Junshan Leng

Fadel Adib

Yunfei Ma

Unsoo Ha

Tzu Ming Hsu

Alaa Khaddaj

Zexuan Zhong

Tara Boroushaki

A control system and method locates a partially or fully occluded target object in an area of interest. The location of the occluded object may be determined using visual information from a vision sensor and RF-based location information. Determining the location of the target object in this manner may effectively allow the control system to 'see through' obstructions that are occluding the object. Model-based and/or deep-learning techniques may then be employed to move a robot into range relative to the target object to perform a predetermined (e.g., grasping) operation. This operation may be performed while the object is still in the occluded state or after a decluttering operation has been performed to remove one or more obstructions that are occluding light-of-sight vision to the object.

System and method for location determination and robot control

A system may sense the contents of a closed container, by analyzing a wireless signal that reflects from an RFID tag on the outside of the container. The frequency response of the tag's antenna may be affected by the relative permittivity of the contents and by the tag's environment. The frequency response may be measured in a line-of-sight environment and in a multipath environment. Channel estimates may be calculated, based on the measurements. Channel ratios may be calculated by dividing line-of-sight channel estimates by multipath channel estimates. The resulting channel ratios may be fed into a variational autoencoder, which in turn generates synthetic data that contains information about multipath environments but not the contents. The output of the variational autoencoder may be converted into synthetic channel estimates, which may in turn be employed for anomaly detection, or to train a classifier to classify contents of the container.

Methods and apparatus for radio frequency sensing in diverse environments

Growing community of inventors

Malden, MA, United States of America

Junshan Leng