The aim of the project was to design a system of simple robots that would act asynchronously and collaborate to work as a limb and manipulate a target. The project was an adaptation of the classic problem collaboration between multiple simple robotic systems to accomplish a complex task. In this case the robots are physically connected and must work directly with each other to complete the goal of manipulating the target. Each robotic "node" has no awareness of the other nodes; however, all nodes can communicate with the master node to provide node specific data. The master node is aware of the overall goal and has sensor input to gauge the success of the entire system. This provides an interesting model that could have the potential to be used in several different situations with little or no input from any human operator. Typical collaborative systems consist of separate vehicles that act together to either map an environment or solve a task such as moving an object. This project would utilize similar principles to create a fairly complex robot by using a few simpler robots. By using these smaller, less complex parts it should be possible to create a system that is highly reconfigurable. This ability would allow a platform like this to be integrated quickly with research platforms as a tool manipulator or in a manufacturing system to handle hazardous materials or complete tedious tasks.
Hardware specs of each node:
- 200 MHz ARM9 processor running Linux
- 32 MB SDRAM
- 8 MB on-board flash
- SD card slot for data logging and program storage
- Custom ZigBee hardware
Texas Instruments Engibous Prize: Poster Session Winner