Autonomous robot navigation and path following via RFID tag detection
One of the main problems associated with Autonomous Robot Navigation is that of positioning. When a robot finds itself in an unknown environment, it must use sensor readings from within the environment to create a map and locate its position. This is known as the Simultaneous Localisation and Mappin...
| Main Author: | |
|---|---|
| Format: | Dissertation (University of Nottingham only) |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/48562/ |
| _version_ | 1848797794617262080 |
|---|---|
| author | Castle-Green, Simon |
| author_facet | Castle-Green, Simon |
| author_sort | Castle-Green, Simon |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | One of the main problems associated with Autonomous Robot Navigation is that of positioning. When a robot finds itself in an unknown environment, it must use sensor readings from within the environment to create a map and locate its position. This is known as the Simultaneous Localisation and Mapping (SLAM) building conundrum. A number of technologies exist in an attempt to solve the SLAM conundrum but can prove expensive and complicated to deploy. This paper looks at utilising Radio Frequency Identification (RFID) technology for the purpose of providing a mechanism in which a robot can follow a path of pre-deployed RFID transponders without any other knowledge of the environment and without utilising other sensors. RFID sys-tems provide additional data called Received Signal Strength Indicators (RSSI) which indicates the strength of the received signal from the RFID Tag. A prototype rotating RFID reader was created allowing for the collection of RSSI data whilst reading an RFID transponder. From the data gathered, a series of Random Forest machine learning models were built in order to allow the prediction of a transponder’s angle in relation to the RFID reader. The resulting models were evaluated within a Java simulator by creating a number of experiment paths. Each exper¬iment contains 20 pre-deployed tags, which a virtual robot armed with a Virtual RFID reader attempts to follow. Simulations were evaluated based on completeness, angle prediction accu¬racy and path deviation. The research showed some success with one of the models achieving an 97% accuracy in angle prediction and successfully completing 95% of the test paths with a low path deviation score. This research shows strong feasibility for a system such as this being used as a means of robot navigation through following RFID Tags. |
| first_indexed | 2025-11-14T20:09:32Z |
| format | Dissertation (University of Nottingham only) |
| id | nottingham-48562 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:09:32Z |
| publishDate | 2017 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-485622018-01-09T14:16:10Z https://eprints.nottingham.ac.uk/48562/ Autonomous robot navigation and path following via RFID tag detection Castle-Green, Simon One of the main problems associated with Autonomous Robot Navigation is that of positioning. When a robot finds itself in an unknown environment, it must use sensor readings from within the environment to create a map and locate its position. This is known as the Simultaneous Localisation and Mapping (SLAM) building conundrum. A number of technologies exist in an attempt to solve the SLAM conundrum but can prove expensive and complicated to deploy. This paper looks at utilising Radio Frequency Identification (RFID) technology for the purpose of providing a mechanism in which a robot can follow a path of pre-deployed RFID transponders without any other knowledge of the environment and without utilising other sensors. RFID sys-tems provide additional data called Received Signal Strength Indicators (RSSI) which indicates the strength of the received signal from the RFID Tag. A prototype rotating RFID reader was created allowing for the collection of RSSI data whilst reading an RFID transponder. From the data gathered, a series of Random Forest machine learning models were built in order to allow the prediction of a transponder’s angle in relation to the RFID reader. The resulting models were evaluated within a Java simulator by creating a number of experiment paths. Each exper¬iment contains 20 pre-deployed tags, which a virtual robot armed with a Virtual RFID reader attempts to follow. Simulations were evaluated based on completeness, angle prediction accu¬racy and path deviation. The research showed some success with one of the models achieving an 97% accuracy in angle prediction and successfully completing 95% of the test paths with a low path deviation score. This research shows strong feasibility for a system such as this being used as a means of robot navigation through following RFID Tags. 2017-12-14 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/48562/1/SimonCastleGreen_MScDissertation.pdf Castle-Green, Simon (2017) Autonomous robot navigation and path following via RFID tag detection. [Dissertation (University of Nottingham only)] Arduino Autonomous Robot Navigation Java Simulator Robot Path Following Radio Frequency Identification Random Forests Received Signal Strength Indicator RFID RFID Tag RSSI Robot Simulator Simultaneous Localisation and Mapping SLAM SparkFun Simultaneous RFID Reader Transponder. |
| spellingShingle | Arduino Autonomous Robot Navigation Java Simulator Robot Path Following Radio Frequency Identification Random Forests Received Signal Strength Indicator RFID RFID Tag RSSI Robot Simulator Simultaneous Localisation and Mapping SLAM SparkFun Simultaneous RFID Reader Transponder. Castle-Green, Simon Autonomous robot navigation and path following via RFID tag detection |
| title | Autonomous robot navigation and path following via
RFID tag detection |
| title_full | Autonomous robot navigation and path following via
RFID tag detection |
| title_fullStr | Autonomous robot navigation and path following via
RFID tag detection |
| title_full_unstemmed | Autonomous robot navigation and path following via
RFID tag detection |
| title_short | Autonomous robot navigation and path following via
RFID tag detection |
| title_sort | autonomous robot navigation and path following via
rfid tag detection |
| topic | Arduino Autonomous Robot Navigation Java Simulator Robot Path Following Radio Frequency Identification Random Forests Received Signal Strength Indicator RFID RFID Tag RSSI Robot Simulator Simultaneous Localisation and Mapping SLAM SparkFun Simultaneous RFID Reader Transponder. |
| url | https://eprints.nottingham.ac.uk/48562/ |