# Loop Closure Detection in Closed Environments

**Authors:** Nils Rottmann, Ralf Bruder, Achim Schweikard, Elmar Rueckert  
**arXiv:** [1908.04558](https://arxiv.org/abs/1908.04558)  
**Date:** 2019-08-13  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1908.04558v1/

## Key points

- Low cost robots often use simplistic and random navigation policies
- Sophisticated mapping and planning approaches require additional sensors
- "Loop Closure Detection in Closed Environments" proposes a novel loop closure detection method based only on odometry data
- The approach can be used with low-range or binary signal sensors together with simple wall following techniques
- The detected loop closing constraints can be included into a pose graph formulation for map estimation
- The approach generates accurate map estimates on the basis of odometry data only, even under large odometry noise
- Improved map estimates enable smart navigation policies in low cost robots to perform goal-directed behavior without requiring expensive sensor technology
- This research has significant implications for the development of more efficient and cost-effective robotic systems that can navigate closed environments with accuracy and reliability.

## Summary

In the field of robotics, low cost robots such as vacuum cleaners or lawn mowers often employ simplistic and random navigation policies. While sophisticated mapping and planning approaches exist, they require additional sensors like LIDAR sensors, cameras or time of flight sensors. In this work titled "Loop Closure Detection in Closed Environments", authors Nils Rottmann, Ralf Bruder, Achim Schweikard and Elmar Rueckert propose a novel loop closure detection method based only on odometry data which can be generated using low-range or binary signal sensors together with simple wall following techniques. The authors demonstrate how to include the detected loop closing constraints into a pose graph formulation such that standard pose graph optimization techniques can be used for map estimation. They evaluate their map estimate and loop closure approach using both simulation and a real lawn mower in complex and realistic environments. The results show that their approach generates accurate map estimates on the basis of odometry data only. The authors further demonstrate that their assumption about the discriminative nature of neighboring poses in the pose graph is solid even under large odometry noise. The improved map estimates provided by this approach enable smart navigation policies in low cost robots to perform goal-directed behavior like pick-and-place or complete coverage path planning in realistic environments without requiring expensive sensor technology. This research presents an innovative solution to a long-standing problem in robotics and has important implications for future developments in this field. Overall, this work has significant implications for the development of more efficient and cost-effective robotic systems that can navigate closed environments with accuracy and reliability.