Reusable Work Primitives for Semi-Autonomous Multi-Task Robots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robots require elaborate tele-operation systems and significant training before deployment, limiting accessibility and efficiency in performing multiple work objectives.
Innovation Solution
A robot system that utilizes a catalog of reusable work primitives, enabling semi-autonomous operation by identifying and executing workflows for various work objectives through a processor and non-transitory storage medium, allowing for switching between different work tasks using a library of reusable primitives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If tele-operation systems are used to control robots, then the robot can perform physical actions emulating a human operator, but the system requires very elaborate and complicated interfaces with sophisticated sensors and equipment, requiring full attention of the pilot and limiting accessibility
Solution Approach 1:
The robot is trained to autonomously perform tasks through repeated practice in the real world, enabling it to serve itself without requiring complex tele-operation systems. The robot develops its own capability to complete work objectives independently, reducing reliance on elaborate control interfaces and pilot attention.
Solution Approach 2:
The patent replaces the mechanical tele-operation control system with a learning-based autonomous system. Instead of using sophisticated sensors and equipment to control the robot remotely, the system uses machine learning models trained on real-world data to enable the robot to autonomously perform tasks, substituting the mechanical control architecture with an intelligent software-based approach.
2Extent of automation
If a robot is trained to operate semi-autonomously or fully autonomously, then the robot can perform tasks without tele-operation, but training involves causing the robot to repeatedly perform physical tasks in the real world, causing significant wear and tear on components before deployment
Solution Approach 1:
The patent uses simulation environments to create virtual copies of the real world for training the robot. Instead of repeatedly performing physical tasks that cause wear and tear, the robot is trained in simulated environments that replicate real-world conditions. The training data generated from these simulations is then used to train machine learning models that enable autonomous operation, eliminating the need for physical wear during the training phase.
3Adaptability or versatility
If a robot is trained by repeatedly performing physical tasks in the real world, then the robot can learn to complete work objectives, but this process causes significant wear and tear on robot components before the robot can be deployed
Solution Approach 1:
The patent creates virtual copies of real-world environments and tasks through simulation. The robot learns task adaptability and versatility by training in these simulated environments rather than physically performing tasks. This approach maintains component reliability while still enabling the robot to learn diverse work objectives through repeated exposure to simulated task variations.
Solution Approach 2:
The patent performs preliminary training actions in simulated environments before actual deployment. By pre-training the robot in virtual simulations that replicate real-world conditions, the system prepares the robot for autonomous task performance without causing wear and tear on physical components during the learning phase.
Data Source
AI summary
Robots, systems, methods, and computer program products for training and operating (semi-)autonomous robots to complete work objectives are described. A robot accesses a library of reusable work primitives from a catalog of libraries of reusable work primitives, each reusable work primitive corresponding to a respective basic sub-action that the robot is trained to autonomously perform. A work objective is analyzed to determine a sequence of reusable work primitives that complete the work objective, and the robot executes the sequence to complete the work objective. A robot can be deployed with access to an appropriate library of reusable work primitives, based on expectations for the robot. The robot is trained to perform reusable work primitives in multiple libraries, by generating control instructions which cause the robot to perform each reusable work primitive. Training is performed by real-world robots performing reusable work primitives, or simulated robot instances performing the reusable work primitives.