Cloud AMR Landmark Control for Indoor-Outdoor Delivery Transitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing autonomous mobile robots (AMR) for both indoor and outdoor environments separately is costly, and there is a need for efficient transition between these environments, especially for delivering heavy objects.
Innovation Solution
A cloud server communicates with the AMR to generate landmarks for indoor and outdoor movement algorithms, predict the transition point, and set additional landmarks based on delivery product attributes, optimizing algorithm usage and delivery accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate AMR robots are developed for indoor and outdoor environments, then each robot can be optimized for its specific environment, but the development cost becomes enormous
Solution Approach 1:
The patent implements a unified AMR platform that can operate in both indoor and outdoor environments by using multiple landmark types (natural landmarks for outdoor, artificial landmarks for indoor) and a single integrated algorithm suite. The system selectively activates appropriate algorithms based on the detected environment type, eliminating the need for separate robot developments while maintaining environment-specific optimization capabilities.
2Reliability
If all movement algorithms are continuously applied during transition, then the robot can handle all possible scenarios, but unnecessary algorithm use increases computational overhead
Solution Approach 1:
The system dynamically selects and activates movement algorithms based on real-time landmark detection and environment classification. When transitioning from outdoor to indoor environments, the system detects the change in landmark types and selectively switches from outdoor-oriented algorithms to indoor-oriented algorithms, ensuring appropriate scenario coverage while minimizing unnecessary computational overhead.
3Use of energy by moving object
If the robot uses only outdoor movement algorithm during outdoor delivery, then energy consumption is reduced, but the robot cannot accurately perform indoor final unloading
Solution Approach 1:
The delivery process is segmented into distinct phases (outdoor delivery to building entrance, indoor navigation to final destination) with corresponding algorithm selections. The system maintains low energy consumption by using outdoor algorithms during outdoor phases and activates indoor algorithms only when transitioning to indoor environments, ensuring both energy efficiency and adaptability to different delivery scenarios.
4Speed
If the system pre-sets all possible landmarks in advance, then algorithm selection is faster, but the system cannot adapt to unexpected environments or changes
Solution Approach 1:
The system employs real-time landmark detection and automatic environment classification to dynamically determine the appropriate movement algorithms. Instead of relying on pre-set landmarks, the robot autonomously identifies environmental features, classifies the current environment type, and selects corresponding algorithms on-the-fly, ensuring both rapid response and adaptability to unexpected environments.
Data Source
AI summary
The present invention relates to a control method of a cloud server capable of communicating with an autonomous mobile robot (AMR), wherein the control method comprises the steps of: receiving, from a mobile phone or a delivery service-related server, address information, delivery product information, and at least one of a drop location-related image or a capture image; generating a first landmark on the basis of the received address information; and generating a second landmark on the basis of the received delivery product information.