Tesla, Inc.

Anduril Industries, Inc.

Forrest Nelson Iandola

Harsimran Singh Sidhu

Paras Jagdish Jain

Anting Shen

Adnan Esmail

Kevin Michael Christensen

Donald Benton MacMillen

Nadav Ben-Haim

Gokul Ganesh Subramanian

Corey Martin Hass

Calvin Michael Hareng

Brian Marcin Camacho

Daniel Paden Tomasello

Yiqi Hou

Rohan Nandkumar Phadte

A neural network architecture is used that reduces the processing load of implementing the neural network. This network architecture may thus be used for reduced-bit processing devices. The architecture may limit the number of bits used for processing and reduce processing to prevent data overflow at individual calculations of the neural network. To implement this architecture, the number of bits used to represent inputs at levels of the network and the related filter masks may also be modified to ensure the number of bits of the output does not overflow the resulting capacity of the reduced-bit processor. To additionally reduce the load for such a network, the network may implement a 'starconv' structure that permits the incorporation of nearby nodes in a layer to balance processing requirements and permit the network to learn from context of other nodes.

Neural networks for embedded devices

A system for automatically maintaining focus while tracking remote flying objects includes an interface and processor. The interface is configured to receive two or more images. The processor is configured to determine a bounding box for an object in the two or more images; determine an estimated position for the object in a future image; and determine an estimated focus setting and an estimated pointing direction for a lens system.

Auto-focus tracking for remote flying targets

A system for automatically acquiring focus of remote flying objects (RFOs) includes an interface and processor. The interface is configured to receive a radar data and a lens temperature data. The processor is configured to determine a focal setting for a lens system based at least in part on the radar data and the lens temperature data; and provide the focal setting for the lens system.

Auto-focus acquisition for remote flying targets

A model training and implementation pipeline trains models for individual embedded systems. The pipeline iterates through multiple models and estimates the performance of the models. During a model generation stage, the pipeline translates the description of the model together with the model parameters into an intermediate representation in a language that is compatible with a virtual machine. The intermediate representation is agnostic or independent to the configuration of the target platform. During a model performance estimation stage, the pipeline evaluates the performance of the models without training the models. Based on the analysis of the performance of the untrained models, a subset of models is selected. The selected models are then trained and the performance of the trained models are analyzed. Based on the analysis of the performance of the trained models, a single model is selected for deployment to the target platform.

Optimizing neural network structures for embedded systems

An autonomous control system generates synthetic data that reflect simulated environments. Specifically, the synthetic data is a representation of sensor data of the simulated environment from the perspective of one or more sensors. The system generates synthetic data by introducing one or more simulated modifications to sensor data captured by the sensors or by simulating the sensor data for a virtual environment. The autonomous control system uses the synthetic data to train computer models for various detection and control algorithms. In general, this allows autonomous control systems to augment training data to improve performance of computer models, simulate scenarios that are not included in existing training data, and/or train computer models that remove unwanted effects or occlusions from sensor data of the environment.

Data synthesis for autonomous control systems

An autonomous control system combines sensor data from multiple sensors to simulate sensor data from high-capacity sensors. The sensor data contains information related to physical environments surrounding vehicles for autonomous guidance. For example, the sensor data may be in the form of images that visually capture scenes of the surrounding environment, geo-location of the vehicles, and the like. The autonomous control system simulates high-capacity sensor data of the physical environment from replacement sensors that may each have lower capacity than high-capacity sensors. The high-capacity sensor data may be simulated via one or more neural network models. The autonomous control system performs various detection and control algorithms on the simulated sensor data to guide the vehicle autonomously.

Multi-channel sensor simulation for autonomous control systems

An annotation system provides various tools for facilitating training data annotation. The annotation tools include a bidirectional annotation model that generates annotations for an image sequence based on both forward information and backward information in an image sequence. The annotation system also facilitates annotation processes by automatically suggesting annotations to the human operator based on a set of annotation predictions and locations of interactions of the human operator on the image. This way, the annotation system provides an accelerated way to generate high-quality annotations that take into account input from a human operator by using the predictions as a guide when it appears that an estimated annotation is consistent with the judgement of the human operator. The annotation system also updates annotations for an overlapping set of objects based on input from human operators.

Growing community of inventors

San Jose, CA, United States of America

Forrest Nelson Iandola