US Patent Application 17732401. VEHICLE TAILLIGHT RECOGNITION BASED ON A VELOCITY ESTIMATION simplified abstract

From WikiPatents
Jump to navigation Jump to search

VEHICLE TAILLIGHT RECOGNITION BASED ON A VELOCITY ESTIMATION

Organization Name

TOYOTA JIDOSHA KABUSHIKI KAISHA


Inventor(s)

Kuan-Hui Lee of San Jose CA (US)

Charles Christopher Ochoa of San Francisco CA (US)

Arjun Bhargava of San Francisco CA (US)

Chao Fang of Sunnyvale CA (US)

VEHICLE TAILLIGHT RECOGNITION BASED ON A VELOCITY ESTIMATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 17732401 titled 'VEHICLE TAILLIGHT RECOGNITION BASED ON A VELOCITY ESTIMATION

Simplified Explanation

The patent application describes a method for controlling a vehicle in a specific environment using a taillight recognition system.

  • The method involves associating objects in the environment with a velocity instance label using a velocity model.
  • A recurrent network of the taillight recognition system selectively focuses on a chosen region of images using a spatial attention model for taillight recognition.
  • The selected region is combined with the velocity instance label of each object to create a concatenated region label.
  • The ego vehicle's trajectory is planned based on the inferred intent of each object, determined by the taillight state of each object as indicated by the concatenated region label.


Original Abstract Submitted

A method for controlling an ego vehicle in an environment includes associating, by a velocity model, one or more objects within the environment with a respective velocity instance label. The method also includes selectively, by a recurrent network of the taillight recognition system, focusing on a selected region of the sequence of images according to a spatial attention model for a vehicle taillight recognition task. The method further includes concatenating the selected region with the respective velocity instance label of each object of the one or more objects within the environment to generate a concatenated region label. The method still further planning a trajectory of the ego vehicle based on inferring, at a classifier of the taillight recognition system, an intent of each object of the one or more objects according to a respective taillight state of each object, as determined based on the concatenated region label.