Multi-Robot Route Planning With Order Slots in Confined Spaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-robot route planning systems struggle to scale effectively in confined spaces, leading to cumbersome re-planning and interference issues when new robots are added, resulting in combinatorial problems and inefficiencies.
Innovation Solution
A processor-based method and system for multi-robot route planning that determines route plans using an order value associated with each node, optimizing routes based on distance and wait time estimates, and dynamically re-evaluating paths to avoid collisions and deadlocks, utilizing a decentralized architecture for efficient communication and path synchronization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional route planning systems are used for multi-robot systems, then robots can navigate in simple environments, but the system cannot scale effectively when new robots are added due to combinatorial explosion of input space
Solution Approach 1:
The patent segments the route planning problem by introducing discrete order slots at each node that robots can occupy. Instead of planning routes for all robots simultaneously (combinatorial approach), the system divides the problem into individual robot decisions about which order slot to occupy at each node, transforming an intractable global optimization problem into manageable local decisions.
Solution Approach 2:
The patent changes the parameter space by introducing order values and order slots as discrete parameters. Robots bid for order slots based on order values (distance metrics), converting continuous route planning into discrete parameter selection. This parameter transformation enables scalable computation by limiting the decision space to a finite set of order slots rather than continuous path possibilities.
2Ease of operation
If robots are forced to follow predetermined lines in confined spaces, then route planning is simplified, but the system cannot handle dynamic obstacles and reconfiguration becomes challenging
Solution Approach 1:
The patent introduces dynamics by allowing robots to bid for and change their order slots at each node based on real-time conditions. The order values and slot assignments are not fixed but dynamically adjusted as robots arrive at nodes and compete for slots, enabling the system to adapt to changing environments and obstacles while maintaining structured route planning.
Solution Approach 2:
The system implements feedback through the bidding mechanism where robots evaluate order values (distance to destination) and bid for order slots based on current system state. This feedback loop allows robots to adjust their routes dynamically, responding to obstacles and other robots' positions while maintaining overall system coordination through the order slot framework.
3Productivity
If existing route planning systems handle multiple robots, then basic navigation is achieved, but re-planning becomes cumbersome and interference issues arise when robots need to adapt to new conditions
Solution Approach 1:
The patent applies preliminary action by pre-defining order slots at each node before robots arrive. This preliminary structure of available slots and order values allows robots to make immediate decisions upon arrival without requiring complex real-time replanning, reducing re-planning time while maintaining navigation efficiency through the pre-established bidding framework.
Data Source
AI summary
The disclosure provides a method and system for multi-robot route planning. The method includes determining a route plan of a node based on an order value associated with the node, wherein the order value is distance travelled from the node to one or more nodes in a network, occupying an order slot from a list of orders of the node. The determined route plan of the node is sent to the one or more nodes in the network and a new route plan is generated, based on order value threshold and wait time estimate associated with the node. The method includes optimizing the generated new route plan of the node comprising computing a new order value occupying a new order slot from the list of orders of the node in parallel to change in status of order value of the one or more nodes.