Waymo LLC

Pei Sun

Dragomir Dimitrov Anguelov

Yin Zhou

Vijay Vasudevan

Jiquan Ngiam

Yuning Chai

Jiyang Gao

Weiyue Wang

Yu Ouyang

Yu Zhang

Jared Stephen Russell

Jens-Steffen Ralf Gutmann

Xiao Meng Zhang

Zijian Guo

Benjamin James Caine

Zhifeng Chen

Alexander Patrick McCauley

Jonathon Shlens

Patrick An Phu Nguyen

Mingxing Tan

Ouais Alsharif

Dumitru Erhan

Benjamin Sapp

Wei Han

Hang Zhao

Chenxi Liu

Henrik Kretzschmar

Xin Huang

Sean Rafferty

Ruizhongtai Qi

Brandon Chauloon Yang

Fei Xia

Zhaoqi Leng

Kyriacos Christoforos Shiarlis

Zhenpei Yang

Xi Yi

Methods, systems, and apparatus for processing point clouds using neural networks to perform a machine learning task. In one aspect, a system comprises one or more computers configured to obtain a set of point clouds captured by one or more sensors. Each point cloud includes a respective plurality of three-dimensional points. The one or more computers assign the three-dimensional points to respective voxels in a voxel grid, where the grid of voxels includes non-empty voxels to which one or more points are assigned and empty voxels to which no points are assigned. For each non-empty voxel, the one or more computers generate initial features based on the points that are assigned to the non-empty voxel. The one or more computers generate multi-scale features of the voxel grid, and the one or more computers generate an output for a point cloud processing task using the multi-scale features of the voxel grid.

Performing point cloud tasks using multi-scale features generated through self-attention

The technology relates controlling an autonomous vehicle through a multi-lane turn. In one example, data corresponding to a position of the autonomous vehicle in a lane of the multi-lane turn, a trajectory of the autonomous vehicle, and data corresponding to positions of objects in a vicinity of the autonomous vehicle may be received. A determination of whether the autonomous vehicle is positioned as a first vehicle in the lane or positioned behind another vehicle in the lane may be made based on a position of the autonomous vehicle in the lane relative to the positions of the objects. The trajectory of the autonomous vehicle through the lane may be adjusted based on whether the autonomous vehicle is positioned as a first vehicle in the lane or positioned behind another vehicle in the lane. The autonomous vehicle may be controlled based on the adjusted trajectory.

Controlling vehicles through multi-lane turns

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generated simulated sensor data. One of the methods includes obtaining a surfel map generated from sensor observations of a real-world environment and generating, for each surfel in the surfel map, a respective grid having a plurality of grid cells, wherein each grid has an orientation matching an orientation of a corresponding surfel, and wherein each grid cell within each grid is assigned a respective color value. For a simulated location within a simulated representation of the real-world environment, a textured surfel rendering is generated, including combining color information from grid cells visible from the simulated location within the simulated representation of the real-world environment.

Generating environmental data

Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for performing object detection. The system obtains a respective range image corresponding to each point cloud in a set of point clouds captured by one or more sensors. The system processes each range image using a segmentation neural network to generate range image features and a segmentation output. The system generates a feature representation of the set of point clouds from only the feature representations of the foreground points. The system processes the feature representation of the set of point clouds using a prediction neural network to generate a prediction characterizing the set of point clouds.

Efficient three-dimensional object detection from point clouds

Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for performing three-dimensional auto-labeling on sensor data. The system obtains a sensor data segment that includes a temporal sequence of three-dimensional point clouds generated from sensor readings of an environment by one or more sensors. The system identifies, from the sensor data segment, (i) a plurality of object tracks that each corresponds to a different object in the environment and (ii) for each object track, respective initial three-dimensional regions in each of one or more of the point clouds in which the corresponding object appears. The system generates, for each object track, extracted object track data that includes at least the points in the respective initial three-dimensional regions for the object track. The system further generates, for each object track and from the extracted object track data for the object track, an auto labeling output that defines respective refined three-dimensional regions in each of the one or more point clouds.

Object-centric three-dimensional auto labeling of point cloud data

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for agent trajectory prediction using temporal-spatial interaction predictions.

Multi-modal multi-agent trajectory prediction

Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for processing a perspective view range image generated from sensor measurements of an environment. The perspective view range image includes a plurality of pixels arranged in a two-dimensional grid and including, for each pixel, (i) features of one or more sensor measurements at a location in the environment corresponding to the pixel and (ii) geometry information comprising range features characterizing a range of the location in the environment corresponding to the pixel relative to the one or more sensors. The system processes the perspective view range image using a first neural network to generate an output feature representation. The first neural network comprises a first perspective point-set aggregation layer comprising a geometry-dependent kernel.

Processing perspective view range images using neural networks

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing point cloud data using dynamic voxelization. When deployed within an on-board system of a vehicle, processing the point cloud data using dynamic voxelization can be used to make autonomous driving decisions for the vehicle with enhanced accuracy, for example by combining representations of point cloud data characterizing a scene from multiple views of the scene.

Processing point clouds using dynamic voxelization

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing point cloud data representing a sensor measurement of a scene captured by one or more sensors to generate an object detection output that identifies locations of one or more objects in the scene. When deployed within an on-board system of a vehicle, the object detection output that is generated can be used to make autonomous driving decisions for the vehicle with enhanced accuracy.

Object detection in point clouds

Growing community of inventors

Palo Alto, CA, United States of America

Pei Sun