Robot Map Motion Primitives With Constraint Regions for Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic devices face challenges in navigating constrained environments while avoiding obstacles and maintaining safe, efficient routes, as conventional navigation systems lack the necessary context to make informed decisions about deviations from the planned path.
Innovation Solution
A navigation system that utilizes a route specification language incorporating context and constraints, such as yaw and position, to enable local navigation systems to make optimized decisions, ensuring safe and efficient traversal through environments with dynamic obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional navigation systems are used, then the robot can navigate to the destination, but the robot lacks context to make informed decisions about path deviations and may contact obstacles or become stuck
Solution Approach 1:
The navigation system is segmented into multiple hierarchical levels: high-level route specification (global path planning) and local navigation (local path execution). Each level operates with different degrees of freedom and context, allowing the robot to maintain global goals while adapting locally to obstacles and environmental constraints without losing overall navigation context.
Solution Approach 2:
The patent implements a nested structure where local navigation systems are embedded within the broader route specification framework. The local navigation operates within constraints defined by the high-level route, and can request modifications when encountering obstacles. This nested architecture allows context to be preserved at multiple levels simultaneously.
2Productivity
If the robot follows a strict planned path, then the navigation route is efficient, but the robot cannot adapt to dynamic obstacles or environmental changes
Solution Approach 1:
The route specification system is designed to be dynamic rather than static. The high-level route can be modified when local navigation encounters obstacles or environmental changes. The system allows for real-time adjustments while maintaining the overall efficiency of the original path, balancing adherence to the planned route with adaptability to changing conditions.
Solution Approach 2:
The local navigation system continuously monitors the environment and provides feedback to the high-level route specification. When obstacles or changes are detected, this feedback triggers route modifications. The system learns from environmental interactions and adjusts the path accordingly, maintaining efficiency while adapting to dynamic conditions.
3Reliability
If the robot navigates through constrained environments with many obstacles, then the robot can reach the destination, but the navigation becomes complex and computationally intensive
Solution Approach 1:
The navigation problem in constrained environments is segmented into manageable components: global route specification identifies key waypoints and corridors, while local navigation handles immediate obstacle avoidance. This segmentation reduces computational complexity by distributing the navigation task across different spatial and temporal scales.
Solution Approach 2:
Different parts of the navigation system operate with different levels of detail and computational intensity. The high-level route specification uses simplified environmental models for overall path planning, while local navigation employs more detailed sensing and computation only when needed for immediate obstacle avoidance. This local quality approach optimizes computational resources.
Data Source
AI summary
A method includes receiving sensor data of an environment about a robot and generating a plurality of waypoints and a plurality of edges each connecting a pair of the waypoints. The method includes receiving a target destination for the robot to navigate to and determining a route specification based on waypoints and corresponding edges for the robot to follow for navigating the robot to the target destination selected from waypoints and edges previously generated. For each waypoint, the method includes generating a goal region encompassing the corresponding waypoint and generating at least one constraint region encompassing a goal region. The at least one constraint region establishes boundaries for the robot to remain within while traversing toward the target destination. The method includes navigating the robot to the target destination by traversing the robot through each goal region while maintaining the robot within the at least one constraint region.