Robot Pose Mapping with Repositionable Initialization Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems face challenges in efficiently navigating and cleaning environments with obstacles, as they often require precise mapping and path planning, and existing methods may require fixed initialization objects or frequent updates when these objects are moved.
Innovation Solution
The system uses repositionable initialization objects to determine the robot's pose and generate maps of the environment, allowing for autonomous or semi-autonomous navigation and cleaning by projecting training trajectories and sensor data onto a grid map, expanding paths based on unoccupied areas, and adjusting paths to avoid obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fixed initialization objects are used for robot pose determination, then mapping precision is improved, but adaptability deteriorates when objects are moved
Solution Approach 1:
The system transitions from static initialization objects to dynamic initialization objects that can be repositioned. The robot determines its pose relative to these objects, and when objects are moved, the system automatically updates the map coordinates through coordinate transformation, allowing the initialization objects to serve multiple positions without requiring system reconfiguration
Solution Approach 2:
The system changes the coordinate system parameters dynamically. When an initialization object is repositioned, the map coordinates are transformed based on the new object position while maintaining the same physical location in the environment. This parameter transformation allows the system to adapt to object movement while preserving mapping precision
2Measurement precision
If frequent map updates are performed when initialization objects move, then mapping accuracy is maintained, but time consumption increases
Solution Approach 1:
The system performs preliminary coordinate transformation calculations based on the known relationship between initialization objects and map coordinates. When an object is repositioned, the pre-established coordinate relationships allow for rapid calculation of new positions without requiring comprehensive remapping, significantly reducing update time while maintaining accuracy
Solution Approach 2:
The system uses virtual copies of the initialization objects in the digital map. When a physical object is repositioned, its virtual counterpart's coordinates are updated through transformation calculations rather than requiring complete remapping. This copying approach maintains mapping accuracy while minimizing the time and computational resources needed for updates
3Reliability
If complex path planning is used to navigate around obstacles, then navigation reliability is improved, but device complexity increases
Solution Approach 1:
The environment is segmented into a grid map with discrete cells, and obstacles are represented as occupied grid cells. The path planning algorithm segments the navigation problem into finding a sequence of valid grid transitions from start to goal, avoiding occupied cells. This segmentation simplifies the complex continuous navigation problem into discrete steps, maintaining reliability while reducing computational complexity
Solution Approach 2:
The grid map serves as an intermediary representation between the physical environment and the path planning algorithm. Instead of directly processing complex sensor data and physical constraints, the system uses the simplified grid map as an intermediary model where path planning can be performed using standard algorithms, reducing overall system complexity while maintaining navigation reliability
Data Source
AI summary
The present disclosure provides systems and methods for robotic navigation, teaching, and mapping. In some cases, the robotic systems and methods may be used to clean an area or environment.