Van Eden, BeatriceBotha, NatashaRosman, B2024-02-052024-02-052023-11Van Eden, B., Botha, N. & Rosman, B. 2023. A comparison of visual place recognition methods using a mobile robot in an indoor environment. http://hdl.handle.net/10204/13561 .http://hdl.handle.net/10204/13561Spatial awareness is an important competence for a mobile robotic system. A robot needs to localise and perform context interpretation to provide any meaningful service. With the deep learning tools and readily available sensors, visual place recognition is a first step towards identifying the environment to bring a robot closer to spatial awareness. In this paper, we implement place recognition on a mobile robot considering a deep learning approach. For simple place classification, where the task involves classifying images into a limited number of categories, all three architectures; VGG16, Inception-v3 and ResNet50, perform well. However, considering the pros and cons, the choice may depend on available computational resources and deployment constraints.FulltextenMobile robotic systemDeep learning toolsVisual place recognitionVGG16ResNet50Inception-v3Conference PresentationVan Eden, B., Botha, N., & Rosman, B. (2023). A comparison of visual place recognition methods using a mobile robot in an indoor environment. http://hdl.handle.net/10204/13561Van Eden, Beatrice, Natasha Botha, and B Rosman. "A comparison of visual place recognition methods using a mobile robot in an indoor environment." <i>RAPDASA-RobMech-PRASA-AMI Conference, CSIR International Convention Centre, Pretoria, South Africa, 30 October – 2 November 2023</i> (2023): http://hdl.handle.net/10204/13561Van Eden B, Botha N, Rosman B, A comparison of visual place recognition methods using a mobile robot in an indoor environment; 2023. http://hdl.handle.net/10204/13561 .TY - Conference Presentation AU - Van Eden, Beatrice AU - Botha, Natasha AU - Rosman, B AB - Spatial awareness is an important competence for a mobile robotic system. A robot needs to localise and perform context interpretation to provide any meaningful service. With the deep learning tools and readily available sensors, visual place recognition is a first step towards identifying the environment to bring a robot closer to spatial awareness. In this paper, we implement place recognition on a mobile robot considering a deep learning approach. For simple place classification, where the task involves classifying images into a limited number of categories, all three architectures; VGG16, Inception-v3 and ResNet50, perform well. However, considering the pros and cons, the choice may depend on available computational resources and deployment constraints. DA - 2023-11 DB - ResearchSpace DP - CSIR J1 - RAPDASA-RobMech-PRASA-AMI Conference, CSIR International Convention Centre, Pretoria, South Africa, 30 October – 2 November 2023 KW - Mobile robotic system KW - Deep learning tools KW - Visual place recognition KW - VGG16 KW - ResNet50 KW - Inception-v3 LK - https://researchspace.csir.co.za PY - 2023 T1 - A comparison of visual place recognition methods using a mobile robot in an indoor environment TI - A comparison of visual place recognition methods using a mobile robot in an indoor environment UR - http://hdl.handle.net/10204/13561 ER -27452