Abstract:The ability to recognize the surface type is crucial for both indoor and outdoor mobile robots. Knowing the surface type can help indoor mobile robots move more safely and adjust their movement accordingly. However, recognizing surface characteristics is challenging since similar planes can appear substantially different; for instance, carpets come in various types and colors. To address this inherent uncertainty in vision-based surface classification, this study first generates a new, unique data set composed of 2,081 surface images (carpet, tiles, and wood) captured in different indoor environments. Secondly, the pre-trained state-of-the-art deep learning models, namely InceptionV3, VGG16, VGG19, ResNet50, Xception, InceptionResNetV2, and MobileNetV2, were utilized to recognize the surface type. Additionally, a lightweight MobileNetV2-modified model was proposed for surface classification. The proposed model has approximately four times fewer total parameters than the original MobileNetV2 model, reducing the size of the trained model weights from 42 MB to 11 MB. Thus, the proposed model can be used in robotic systems with limited computational capacity and embedded systems. Lastly, several optimizers, such as SGD, RMSProp, Adam, Adadelta, Adamax, Adagrad, and Nadam, are applied to distinguish the most efficient network. Experimental results demonstrate that the proposed model outperforms all other applied methods and existing approaches in the literature by achieving 99.52% accuracy and an average score of 99.66% in precision, recall, and F1-score. In addition to this, the proposed lightweight model was tested in real-time on a mobile robot in 11 scenarios consisting of various indoor environments such as offices, hallways, and homes, resulting in an accuracy of 99.25%. Finally, each model was evaluated in terms of model loading time and processing time. The proposed model requires less loading and processing time than the other models.

Surface Classification Based on Vibration on Omni-Wheel Mobile Base

Adaptive and Robust Terrain Classification Control Algorithm for a Spherical Robot

Three-Dimensional Vibration-Based Terrain Classification for Mobile Robots

Binary Classification of Floor Vibrations for Human Activity Detection Based on Dynamic Mode Decomposition.

Vibration-Based Terrain Identification for Planetary Exploration Robots Using Support Vector Machine

Recognition of surface texture with wearable tactile sensor array: A pilot Study

Floor-Cleaning Robot Using Omni-Directional Wheels

Comparative Study of Different Methods in Vibration-Based Terrain Classification for Wheeled Robots with Shock Absorbers

The Design and Development of an Omni-Directional Mobile Robot Oriented to an Intelligent Manufacturing System

Design and Analysis of an Omni-Directional Mobile Robot with Orthogonal Wheel

Rigidity-Based Surface Recognition for a Domestic Legged Robot

Acoustic tactile sensing for mobile robot wheels

Indoor surface classification for mobile robots

Fine-grained Vibration Based Sensing Using a Smartphone

Classification of Floor Materials Using Piezoelectric Actuator–Sensor Pair and Deep Learning for Mobile Robots

Disturbance Analysis And Control For An Omnidirectional Wheeled Mobile Robot

Accurate and Robust Odometry by Fusing Monocular Visual, Inertial, and Wheel Encoder.

Research of Ground Type Classification Algorithm based on Wheel Force Signal

Laplacian Support Vector Machine for Vibration-Based Robotic Terrain Classification

Robust Tightly Coupled Pose Measurement Based on Multi-Sensor Fusion in Mobile Robot System

AI-Enabled Predictive Maintenance Framework for Autonomous Mobile Cleaning Robots