An Unmanned Ground Vehicle (UGV) is a vehicle that operates while in contact with the ground and without an on-board human presence.
An autonomous UGV is a vehicle that performs behaviors or tasks with a high degree of autonomy, at which it can:
gain information about the environment
work for an extended period of time without human intervention
move itself throughout its operating environment without human assistance
avoid situations that are harmful to people, property, or itself
iCab is an acronym for Intelligent Campus Automobile, at which multiple unmanned ground vehicles (electric golf carts) navigate autonomously within the campus vicinity to transport visitors from one spot to another between buildings. Currently there are two golf carts, modified mechanically and electronically in order to achieve the project objective. These modifications include:
deactivating the throttle paddle and apply direct control of the electric engine for the translational motion
removing the steering wheel and replacing it by electric motor-encoder system for autonomous maneuvers
Each golf cart is equipped with multiple sensors for the environment perception. These sensors enable the vehicle to navigate in dynamic surroundings. Accordingly, the vehicle is able to perform the necessary intelligent maneuvers to reach its destination through a collision-free path. Furthermore, the system is able to detect and classify static obstacles (e.g. trees, lamp poles, buildings, generic obstacles), dynamic obstacles (e.g. vehicles, pedestrians, animals, cyclists, motorists), free navigable space among others. The system bases its navigational decisions on global and local maps, using path planning algorithm.
These sensors include:
GPS & Compass Module
Moreover, each vehicle is equipped with a wireless module in order to communicate with each other and share information regarding the scenario, environment, maps and more. The communication among the vehicles is considered one of the key factors in fulfilling the global task of cooperation and coordination of multiple autonomous unmanned ground vehicles.
The control of the vehicles is achieved via on-board embedded computer, which is running Ubuntu operating system. All algorithms are implemented in Robot Operating System (ROS) architecture, thus communication between different processes and multiple sensors data fusion are achieved for better overall performance.
Path Planning and Navigation
Obstacles Detection and Classification
Cooperation and Coordination of Multi-agent System
28th of October 2015 The experiment was performed to show iCab 1 ability to pick-up a passenger from one side of campus building and drop him off at the other side. The vehicle operated autonomously, without human intervention, and achieved the goal in an excellent manner.