Amazon patents AI that helps robots and drones focus on what matters
Amazon has patented a method for training AI models that control autonomous vehicles, robots, or drones. The system uses 'attention layers' to decide which parts of the environment are most important, allowing the machine to choose actions based on those key details.
In the filing’s own words
A data source configured to provide a representation of an environment of one or more agents is identified. Using a data set obtained from the data source, a neural network-based reinforcement learning model with one or more attention layers is trained. Importance indicators generated by the attention layers are used to identify actions to be initiated by an agent. A trained version of the model is stored.
What exactly they claim (claim 1)
1. A computer-implemented method, comprising: deploying, in response to a first programmatic request received at a network-accessible service of a cloud computing environment from a client, a machine learning model to an autonomous agent incorporated within a vehicle, robot, or, drone, the autonomous agent configured to execute the machine learning model, wherein the machine model includes a subcomponent comprising one or more attention layers, wherein the subcomponent comprising one or more attention layers is trained jointly with one or more other sub-components of the machine learning model, and wherein the machine learning model is trained to, when executed by the autonomous agent, initiate actions taken by the autonomous agent based at least in part on analysis of an environment of the autonomous agent; causing to be presented, in response to a second programmatic request received at the network-accessible service from the client, an indication of a first importance assigned to a first portion of an environment of the autonomous agent by one or more attention layers of the machine learning model executing on the autonomous agent, wherein the first importance exceeds a second importance assigned to a second portion of the environment; and causing, by the machine learning model executing on the autonomous agent, the autonomous agent to initiate a particular action comprising a movement of the vehicle, robot or drone, wherein the particular action is selected, by the machine learning model, based at least in part on the indication of the first importance and contents of the first portion of the environment assigned the first importance.
Facts
- Company
- Amazon (Amazon Technologies, Inc.)
- Official title
- Attention-based deep reinforcement learning for autonomous agents
- Patent number
- US12792912B2
- Application number
- 18465916
- Filed
- Sep 12, 2023
- Granted
- Nov 10, 2026
- Inventors
- 4
- Claims
- 14
Classified under
G06N 3/08 Physics & computingG05D 1/0221 Physics & computingG06F 16/904 Physics & computingG06F 17/16 Physics & computingMore from Amazon in AI & software
- Amazon patents system that cites sources for AI answers Sep 29, 2026
- Amazon patents voice assistants that adjust personality based on your mood Nov 3, 2026
- Amazon patents tech to swap words in movies using actor's voice Sep 8, 2026
- Amazon patents ML tool to add graphics to live sports from moving cameras Sep 15, 2026
- Amazon patents shopping assistant that remembers your search filters Oct 6, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.