Robot Intelligence Kernel Dynamic Autonomy Event Horizon
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot architectures lack a scalable framework for dynamic autonomy and generic robot attributes, making it difficult to port behaviors across different robot platforms and applications, resulting in limited intelligence and reliance on human operators for guidance.
Innovation Solution
A robot intelligence kernel with a generic scalable architecture that provides dynamic autonomy and generic robot attributes, enabling seamless porting of behaviors across various platforms by defining an event horizon and adjusting velocity based on obstacle detection, and incorporating a system controller with perceptors and locomotors for environmental perception and mobility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a generic scalable robot architecture is implemented, then adaptability across different robot platforms is improved, but device complexity increases
Solution Approach 1:
The robot architecture is divided into distinct hierarchical layers: hardware abstraction layer, robot abstraction layer, behavior layer, and cognitive layer. Each layer handles specific functions and can be independently developed and ported across different robot platforms, reducing the complexity of implementing adaptability across the entire system.
Solution Approach 2:
The patent creates universal abstraction layers that can interface with multiple different hardware platforms and sensor/actuator types. The hardware abstraction layer and robot abstraction layer provide standardized interfaces that work across diverse robot platforms, allowing the same behavior and cognitive software to operate on different physical systems without modification.
2Extent of automation
If dynamic autonomy framework is implemented, then extent of automation is improved, but device complexity increases
Solution Approach 1:
The architecture implements dynamic autonomy by allowing the robot to transition between different levels of autonomy (fully autonomous, semi-autonomous, teleoperated) based on task requirements and environmental conditions. The behavioral framework dynamically selects and combines behaviors based on sensor inputs and task goals, providing flexible automation without requiring a completely different control system for each autonomy level.
Solution Approach 2:
The system pre-defines a library of reusable behaviors and cognitive functions that can be composed to achieve different autonomy levels. Rather than building autonomy capabilities from scratch for each application, the framework provides pre-built building blocks that can be activated in advance, reducing the complexity of implementing dynamic autonomy.
3Reliability
If event horizon-based velocity control is implemented, then reliability of obstacle avoidance is improved, but device complexity increases
Solution Approach 1:
The event horizon control mechanism continuously monitors sensor data (laser range finders, sonar) to detect obstacles and dynamically adjusts robot velocity based on the distance to detected objects. This real-time feedback loop ensures reliable obstacle avoidance by automatically reducing speed when obstacles are detected within the event horizon distance, while maintaining normal speed when the path is clear.
Solution Approach 2:
The system changes the velocity parameter dynamically based on the event horizon calculation. When obstacles are detected within the event horizon (a calculated distance based on robot speed and stopping distance), the velocity parameter is reduced or set to zero. This parameter adjustment provides simple yet reliable obstacle avoidance without requiring complex control algorithms.
Data Source
AI summary
A robot platform includes perceptors, locomotors, and a system controller. The system controller executes instructions for repeating, on each iteration through an event timing loop, the acts of defining an event horizon, detecting a range to obstacles around the robot, and testing for an event horizon intrusion. Defining the event horizon includes determining a distance from the robot that is proportional to a current velocity of the robot and testing for the event horizon intrusion includes determining if any range to the obstacles is within the event horizon. Finally, on each iteration through the event timing loop, the method includes reducing the current velocity of the robot in proportion to a loop period of the event timing loop if the event horizon intrusion occurs.