Visual Terrain Classification Methods for Mobile Robots Using Hybrid Coding Architecture

Visual terrain classification can provide crucial and important information for motion control and autonomous navigation for mobile robots in complex terrain environment, becoming an important but challenging task. This paper uses a novel hybrid coding architecture, Deep Filter Banks (DFB), combinin...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2019 IEEE 4th International Conference on Image, Vision and Computing (ICIVC) s. 17 - 22
Hlavní autoři: Wu, Hang, Zhang, Wenchang, Li, Boyang, Sun, Yucao, Duan, Deguang, Chen, Ping
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.07.2019
Témata:
arc-legged mobile robot
Autonomous robots
Benchmark testing
Computer architecture
deep filter banks
Encoding
Filter banks
fisher vector
Mobile robots
Motion control
terrain dataset
Vectors
Videos
visual terrain classification
Visualization
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Visual terrain classification can provide crucial and important information for motion control and autonomous navigation for mobile robots in complex terrain environment, becoming an important but challenging task. This paper uses a novel hybrid coding architecture, Deep Filter Banks (DFB), combining stacked denoising sparse autoencoder (SDSAE) and Fisher Vector (FV) for visual terrain classification. Then, we propose a terrain dataset, termed "Terrain8", which is the first publicly available benchmark for visual terrain classification. This dataset contains 2400 terrain images, covering 8 terrain classes with 300 images in each class. Our method achieves superior performance on the Terrain8 dataset. Moreover, we design the framework to deal with terrain videos and carry out the field experiments in arc-legged mobile robot. The field experimental results also indicate the effectiveness of our proposed methods.
DOI:10.1109/ICIVC47709.2019.8981092