Layered Motion Policies for Safe Robot Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in programming machines, such as robots, to perform complex movements safely and predictably, especially in dynamic and uncertain environments.
Innovation Solution
A policy-based control system using a layered approach, where each policy layer generates a movement for the machine, allowing for the combination of multiple layers to achieve complex tasks while ensuring safety and predictability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a machine performs complex tasks with increased programming complexity, then the machine's task capability is improved, but the programming complexity and safety assurance difficulty increase
Solution Approach 1:
The control policy is segmented into multiple hierarchical layers, where each layer handles specific aspects of task execution. The high-level policy handles strategic decision-making while low-level policies handle specific motor control tasks, thereby managing complexity through division of labor across different control levels
Solution Approach 2:
A policy layer acts as an intermediary between high-level task specifications and low-level motor control. Each policy layer processes inputs from higher layers and generates outputs for lower layers, mediating the complex interaction between task goals and physical execution while maintaining safety constraints
2Productivity
If multiple policy layers are combined to achieve complex tasks, then task execution capability is improved, but the system complexity increases
Solution Approach 1:
The control system is divided into segmented policy layers, each responsible for specific control functions. This segmentation allows complex tasks to be broken down into manageable sub-tasks handled by individual layers, improving execution capability while keeping each layer's complexity manageable
Solution Approach 2:
Policy layers are nested hierarchically, with higher-level policies containing and coordinating lower-level policies. Each layer operates within the constraints and context provided by upper layers, creating a nested structure that enables complex behavior through composition of simpler layered components
3Adaptability or versatility
If the machine operates in dynamic and uncertain environments, then adaptability is improved, but safety and predictability become more difficult to ensure
Solution Approach 1:
Safety constraints and operational boundaries are predefined and embedded in the policy layers before execution. Each policy layer is designed with built-in safety checks and predictability guarantees, allowing the system to adapt to dynamic environments while maintaining pre-established safety and reliability properties
Data Source
AI summary
Apparatuses, systems, and techniques provide a policy that can be executed to cause a machine to move. In at least one embodiment, a first policy layer is provided to cause the machine to execute a first motion that causes the machine to accelerate to reach an unbiased state. A second policy layer is provided to cause the machine to execute a second motion without influencing the unbiased state to be reached by machine. The policy can comprise the first and second policy layers.