Semantic Obstacle Planning for Socially Aware Robot Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robots face challenges in navigating environments safely and socially, as existing technologies often fail to effectively differentiate between various obstacles, leading to potential collisions and inadequate consideration of social boundaries.
Innovation Solution
The system employs a planning engine that identifies semantic classes for obstacles, applying specific planning policies to optimize route planning, ensuring collision avoidance and social acceptability by assigning different constraints based on obstacle types, such as buffer zones and speed limits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the robot applies a uniform planning policy to all obstacles, then the device complexity is reduced, but the robot cannot adequately differentiate between various obstacle types leading to inadequate collision avoidance and social boundary consideration
Solution Approach 1:
The patent segments obstacles into different semantic classes (e.g., humans, pets, inanimate objects) and applies distinct planning policies to each class. This segmentation enables the robot to differentiate between obstacle types and apply appropriate buffer zones and constraints, resolving the contradiction by improving adaptability through classification while managing complexity through structured policy organization.
Solution Approach 2:
The patent implements local quality by assigning different planning policy characteristics to different obstacle classes. Each semantic class receives customized planning parameters (e.g., larger buffer zones for humans, smaller for inanimate objects), allowing the robot to adapt its behavior locally to the specific obstacle type encountered rather than applying a uniform global policy.
2Reliability
If the robot maintains large buffer zones around all obstacles, then collision avoidance is improved, but the robot's mobility and task completion efficiency are reduced
Solution Approach 1:
The patent applies local quality by varying buffer zone sizes based on obstacle semantic class. High-risk obstacles like humans receive larger buffer zones for safety, while low-risk obstacles like inanimate objects receive smaller buffer zones, allowing the robot to maintain reliability for critical collisions while improving navigation efficiency by reducing unnecessary spacing for safe obstacles.
Solution Approach 2:
The patent changes the buffer zone parameter dynamically based on the semantic class of the obstacle. The planning policy adjusts spatial parameters (buffer zone radius, minimum approach distance) according to obstacle type, enabling the robot to optimize the balance between collision avoidance reliability and navigation productivity by applying appropriate parameter values to different obstacle categories.
3Measurement precision
If the robot uses detailed semantic classification for obstacles, then the planning precision is improved, but the data processing time and computational latency increase
Solution Approach 1:
The patent segments the classification process into hierarchical levels, first identifying basic obstacle categories (human, pet, object) and then applying corresponding planning policies. This segmentation enables reasonably accurate classification without requiring exhaustive detailed analysis of every obstacle, reducing processing time while maintaining sufficient precision for safe navigation.
Solution Approach 2:
The patent applies partial classification action by identifying only the essential semantic features needed for safe navigation (e.g., distinguishing living from non-living obstacles) rather than performing complete detailed classification. This partial action approach achieves sufficient planning precision for collision avoidance while minimizing computational latency by avoiding unnecessary detailed analysis.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Systems, methods, devices, and techniques for planning travel of an autonomous robot. A system identifies one or more obstacles that are located in proximity of at least a portion of a planned route for the autonomous robot. For each obstacle, the system: (i) determines a semantic class of the obstacle, including selecting the semantic class from a library that defines a set of multiple possible semantic classes for obstacles, and (ii) selects a planning policy for the obstacle that corresponds to the semantic class of the obstacle. The system can generate a trajectory along the at least the portion of the planned route using the selected planning policies. The robot can then initiate travel according to the trajectory.