Custom Search
 

Swarm Robotics

 

 

 

 

         

Swarm Robotics is a new branch in the field of robotics that deals with the behavior and coordination of a large group of robots interacting with each other and with the environment. This multi-robot system approach is based on computer science advances in what is known as 'artificial swarm intelligence', which models and simulates the swarm behavior exhibited by social insects like ants and bees.

   

 

Unlike other multi-robot systems (such as distributed robotic systems), swarm robotics is concerned with the ability to control the behavior of a really large number of robots and provide a high degree of scalability to such systems. Also, just like the biological swarms being emulated by swarm robotics, the individual robots comprising the swarm are not designed to be big and sophisticated - they are physically small with just the ability to carry out simple instructions.

 

But that is the beauty of swarm robotics - individual robots execute simple tasks aimed at interacting with their fellow robots and their environment, but the emerging 'collective' behavior of the entire swarm is what's supposed to accomplish the task that the swarm is deployed for.

  

Experts say that the key to swarm robotics is efficient communication among the swarm robots, which should also provide the robots constant feedback on their group behavior.  This communication system should allow the individual robots to change their respective interactive behaviors whenever necessary to ensure the achievement of the swarm's goal. Wireless communication between the robots may be achieved by radio frequency (RF) or infrared transmission systems.

  

Figure 1.   Photos of Robotic Swarms

   

Needless to say, the approach to programming swarm robots is different from what would be used on an individual robot with centralized intelligence.  Each individual robot in a swarm will be given simple rules on how to interact with the other robots around it, taking into consideration the environment in which they are interacting.  Thus, the behavior of the entire swarm is not being programmed directly.  It is the task of the swarm programmer to provide the correct individual robot instructions that will ultimately result in the desired swarm behavior. 

   

Artificial swarm behavior must be systematically studied, which is why various techniques for achieving this have been developed including video tracking and ultrasonic position monitoring. Software must likewise be engineered very well to ensure that swarm behavior is predicted reliably from individual robot instructions.

  

Michael Crichton's novel, Prey, examines the social and environmental risks of deploying swarm robotics in the outside world without fully understanding artificial swarm behavior, especially when this technology is merged with microbotics (miniature robotics), as it inevitably will be.

  

See a video of a Self-Assembling Robotic Swarm

      

 

   

See Also:   More Industry Articles