Toyota Motor Engineering & Manufacturing North America, Inc.

Toyota Jidosha Kabushiki Kaisha

Avdhut S Joshi

Michael R James

Danil V Prokhorov

Michael Edward Samples

A method of modeling an intersection structure of a roadway. The method includes receiving a first data set including road lane information, and receiving a second data set including vehicle trajectory information for an intersection structure of a roadway. The method includes determining lane node locations from at least one of the first and second data sets. A set of potential links between the lane node locations may be compiled. The method may further include assessing, for each link, a probability that the link is a valid link, and assigning each link with a probability value. The links may be filtered based on a predetermined threshold probability value and a set of valid links is generated. A model of the intersection structure is created based on the set of valid links.

Joint probabilistic modeling and inference of intersection structure

Detecting traffic signaling of a mobile environment includes receiving image frames captured by an imaging device to detect a traffic signal candidate therefrom based on at least a vehicle location. A score having a predefined value is established for the traffic signal candidate. The traffic signal candidate is tracked by incrementing the score when the traffic signal candidate is determined to appear, based on a probabilistic estimator that indicates an expected value for a location of the traffic signal candidate, in a tracked image frame of the captured image frames. The score is decremented when the traffic signal candidate is determined not to appear, based on the probabilistic estimator, in the tracked image frame. The traffic signal candidate is recognized as a traffic signal based on whether the score is above a specified threshold.

System and method for traffic signal recognition

A method for generating accurate lane level maps based on course map information and Lidar data captured during the pass of a sensor carrying vehicle along a road. The method generates accurate lane estimates including the center of each lane, the number of lanes, and the presence of any bicycle paths and entrance and exit ramps using a computer-implemented method where the course map data and the Lidar data are subjected to particle filtering.

Method for generating accurate lane level maps

A vehicle system includes a first sensor and a second sensor, each having, respectively, different first and second modalities. A controller includes a processor configured to: receive a first sensor input from the first sensor and a second sensor input from the second sensor; detect, synchronously, first and second observations from, respectively, the first and second sensor inputs; project the detected first and second observations onto a graph network; associate the first and second observations with a target on the graph network, the target having a trajectory on the graph network; select either the first or the second observation as a best observation based on characteristics of the first and second sensors; and estimate a current position of the target by performing a prediction based on the best observation and a current timestamp.

Tracking on-road vehicles with sensors of different modalities

A method for generating accurate lane level maps based on coarse map information and image sensor data captured during the pass of a sensor carrying vehicle along a road. The method generates accurate lane estimates including any of the center of each lane, the number of lanes, and the presence of any bicycle paths, and entrance and exit ramps, merging of two lanes into one lane, or the expansion of one lane into two lanes, using a computer-implemented method where the coarse map data and the image sensor data are subjected to dual reversible jump-Markov chain Monte Carlo filtering.

A system, device, and methods for autonomous navigation using lane-based localization. Once example computer-implemented method includes receiving, from one or more sensors disposed on a vehicle, data representing a road surface proximate to the vehicle and removing the data falling below an adaptive threshold from the data representing the road surface to isolate data representing boundaries on the road surface. The method further includes generating detected lanes based on the data representing boundaries on the road surface by applying one or more filters, generating expected lanes proximate to the vehicle using data included in a route network definition file, comparing the detected lanes to the expected lanes, and generating a localized vehicle position based on the comparison between the detected lanes and the expected lanes.

Growing community of inventors

Ann Arbor, MI, United States of America

Avdhut S Joshi