Table 3 Comparisons of localization and navigation system of indoor mobile robots
From: Wi-Fi indoor positioning and navigation: a cloudlet-based cloud computing approach
Author | Localization | Cognition | Operation |
|---|---|---|---|
Corderio et al. [47] | Odometry | Line following | A robot can autonomously move following the desired trajectory while avoiding detected obstacles based on depth images |
Bessa et al. [48] | Pattern recognition techniques in omnidirectional images | Artificial neural networks in omnidirectional images | A robot uses pattern recognition techniques in omnidirectional images to estimate the localization of the robot |
Zhang et al. [49] | QR code | Path planning is performed using the Dijkstra algorithm and dynamic window approach | QR codes are used as landmarks to provide global pose references for mobile robot localization and navigation Uses a Laser Ranger Finder (LRF) to avoid collisions |
Mota et al. [50] | Cards and RFID reader | Line following and dynamics of the Petri nets | A robot is equipped with three infrared sensors to detect and follows a black line connecting each card A robot moves until it passes over cards with RFID At each intersection, a robot performs actions, such as turning right or left according to the map defined in its algorithm. Next, it goes straight to the next card |
Our work | A cloudlet-based cloud computing approach | Path planning is made by Dijkstra algorithm and consists of Internet Protocols (IPs) address of APs | A robot is equipped with Raspberry Pi as a wireless access point to connect to APs Cloudlets are deployed at APs A robot moves until it reaches stable segment of AP defined in its path planning At each stable segment of AP, a robot performs actions, such as turning right or left or going straight according to movement decision algorithm |