Multi-Robot SLAM Failover Using Reliability-Based Leader Handover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multi-robot systems face challenges in maintaining reliable simultaneous localization and mapping (SLAM) due to hardware failures and connectivity issues, especially in critical missions where map merging and orienting are resource-intensive and prone to failure if a primary robot fails.

Innovation Solution

A fault-tolerant SLAM system is implemented in robotic clusters, where a leader robot and standby robots calculate reliability and task completion times, allowing for seamless task handover when the leader robot's reliability falls below a threshold, ensuring continuous operation through distributed execution and parallelization of SLAM tasks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If map merging and orienting are performed by a single primary robot, then the SLAM process can be completed, but the system becomes vulnerable to failure if the primary robot fails

Engineering Contradiction:
ImproveSLAM process reliabilityVSAvoidrobotic cluster structure
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the robotic cluster into distinct segments: leader robots that perform map building, member robots that assist, and standby robots that provide backup. This segmentation allows the system to distribute SLAM tasks across multiple robots rather than relying on a single primary robot, thereby improving reliability while maintaining manageable complexity through structured roles.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by pre-positioning standby robots in the cluster before failures occur. These standby robots are prepared in advance with the necessary computational resources and algorithms to take over leader robot functions immediately upon failure detection, eliminating the need for complex real-time reconfiguration and ensuring continuous SLAM operation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple robots perform distributed SLAM, then reliability improves through redundancy, but processing and memory requirements increase significantly

Engineering Contradiction:
ImproveSLAM process reliabilityVSAvoidcomputational resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent merges the computational efforts of multiple robots by having member robots contribute to map building tasks led by leader robots. Instead of each robot independently performing complete SLAM (which would duplicate processing and memory usage), the system combines their computational resources to collectively build and maintain the global map, reducing individual robot resource consumption while maintaining overall system reliability.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements multi-functionality by designing robots that can perform multiple roles depending on system needs. Leader robots can transition to member robot roles when failures occur, and standby robots are designed with universal capabilities to immediately assume leader functions. This flexibility allows the system to maintain reliability through redundancy without requiring each robot to permanently dedicate high computational resources to single functions.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Device complexity

If a single leader robot performs all SLAM tasks, then the system structure is simple, but the system fails completely if the leader robot fails

Engineering Contradiction:
Improvesystem structureVSAvoidSLAM process continuity
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the leader robot's responsibilities by introducing member robots that assist with specific SLAM tasks. The leader robot maintains overall coordination and global map management, while member robots handle localized map building and data processing. This segmentation reduces the computational burden on the leader and creates natural failover points, improving reliability without significantly complicating the overall hierarchical structure.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent prepares standby robots in advance with the knowledge and capability to become new leaders. During normal operation, standby robots monitor the leader's performance and maintain readiness to take over. This preliminary preparation ensures that when leader failure occurs, the system can immediately transition to a new leader without complex real-time decision-making, maintaining simple operational structure while ensuring continuity.

Inventive Principle:
Principle #10Preliminary action

4Use of energy by moving object

If map building is performed by a single robot, then resource requirements are low, but the system cannot tolerate hardware failures

Engineering Contradiction:
Improvecomputational resource consumptionVSAvoidsystem fault tolerance
Core Design Contradiction:
Use of energy by moving objectVSReliability

Solution Approach 1:

The patent merges computational resources across multiple robots performing distributed SLAM. Instead of one robot consuming all computational resources for map building, the system combines the processing power and memory resources of leader and member robots working together. This distributed resource pool provides fault tolerance—if one robot fails, others can continue the task—while keeping individual robot resource consumption manageable.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates copies of critical SLAM data and algorithms across multiple robots in the cluster. Leader and member robots maintain copies of the global map and SLAM processing capabilities, ensuring that if one robot fails, the information and computational ability to continue map building are preserved in other robots. This copying strategy provides redundancy without requiring every robot to maintain full independent SLAM capabilities, balancing resource usage with fault tolerance.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3435190B1System and method for executing fault-tolerant simultaneous localization and mapping in robotic clusters
Publication Date: 2021.10.20 TATA CONSULTANCY SERVICES LTD
  • EP3435190B1 patent drawingFigure 1
  • EP3435190B1 patent drawingFigure 2
  • EP3435190B1 patent drawingFigure 3A

AI summary

In current distributed simultaneous localization and mapping (SLAM) implementations on multiple robots in a robotic cluster, failure of a leader robot terminates a map building process between multiple robots. Therefore, a technique for fault-tolerant SLAM in robotic clusters is disclosed. In this technique, robotic localization and mapping SLAM is executed in a resource constrained robotic cluster such that the distributed SLAM is executed in a reliable fashion and self-healed in case of failure of the leader robot. To ensure fault tolerance, the robots are enabled, by time series analysis, to find their individual failure probabilities and use that to enhance cluster reliability in a distributed manner.