After three years of research (yeah, right!), i've finally begun to write my thesis. So here's the abstract of it:
Title: Towards Load Balanced De-congested Multi Robotic Agent Traffic Control by Coordinated Control at Intersections
Abstract: Network mediated robot navigation has become popular in recent years from different viewpoints. Firstly the network acts as a computing medium thereby reducing the computational payload on-board the robot. In a manner akin to swarm robotics where each of the individual entity has limited intelligence but the group in itself behaves as a sufficiently intelligent system, the network allows the robots to be possessed with minimal decision making capabilities but the network plus the robot behaves as a system of enhanced intelligence. Secondly the network provides for fault tolerance capabilities for if the on-board sensors fail or misbehave the robotic agent can look up to the network for information about the environment. Thirdly the network supplements the computational capacity of the robot. Efficiently designed sensor fusion algorithms can agglomerate intelligence gathered through on-board as well as off board resources to come up with robust decisions. In this work, we developed algorithms for multi robotic traffic control in a world mediated by a network. While single robot navigation mediated by a network is well studied there has been little in the area of multi robotic navigation.
In this work, we present a methodology for coordination of multiple robotic agents moving from one location to another in an environment embedded with a network of agents, placed at strategic locations such as intersections. These intersection agents communicate with robotic agents and also with each other to route robots in a way as to minimize the traffic congestion, thus, resulting in the continuous flow of robots. A robot's path to its destination is computed by the network in terms of the next waypoints to reach. The intersection agents are capable of identifying robots in their proximity based on signal strength. An intersection agent controls the flow of agent traffic around it with the help of the data it collects from the messages received from the robots and other surrounding intersection agents.
The congestion of traffic is minimized using a two layered hierarchical strategy. The objective at the primary level, operating at the intersection, is to reduce congestion at the intersection by reducing the time delay of robots crossing them. The flow of traffic is coordinated by assigning priorities to pathways based on the robot density and its rate of change. In this method the robotic agent hailing from the pathway having the highest agent density and lowest rate of change is allotted the highest priority and the paths of agents with lower priority are attuned to accommodate the paths of the robot with higher priority. The intersection agent allows passage for the robot, that has requested space-time allocation at the intersection, till the point of no conflict in the path of the vehicle through the intersection. The secondary layer maintains coordination between intersection agents and routes traffic such that delay is reduced through effective traffic load balancing. The load balancing of robots over multiple intersections is achieved through coordination between intersection agents by communication of robot densities in different pathways. Routing of a robot from its initial position to its destination in the best available path is done. Individually each robot's path in the environment is distributedly computed by the network as a sequence of waypoints to the goal, each successive waypoint one hop less than the previous. Here, the waypoints are the intersections. Intersection agents coordinate to propagate the robot traffic density information to their neighbors so that at each intersection, the agent guides a robot towards the next waypoint which promises minimum time delay.
In the area of multi-agent traffic control, work has been done based on a first come first served like policy. Our method of coordination at intersections is achieved by considering density and rate of change of density along the incoming pathways to an intersection. Extensive comparisons show the performance gain of the current method over existing ones. Theoretical analysis apart from simulation show the advantages of load balanced traffic flow over uncoordinated allotment of robotic agents to pathways - uniform traffic load is maintained at all intersections and no particular intersection is overwhelmed with robots while others are free of them. Transferring the burden of coordination to the network releases more computational power for the robots to engage in critical assistive activities.
Title: Towards Load Balanced De-congested Multi Robotic Agent Traffic Control by Coordinated Control at Intersections
Abstract: Network mediated robot navigation has become popular in recent years from different viewpoints. Firstly the network acts as a computing medium thereby reducing the computational payload on-board the robot. In a manner akin to swarm robotics where each of the individual entity has limited intelligence but the group in itself behaves as a sufficiently intelligent system, the network allows the robots to be possessed with minimal decision making capabilities but the network plus the robot behaves as a system of enhanced intelligence. Secondly the network provides for fault tolerance capabilities for if the on-board sensors fail or misbehave the robotic agent can look up to the network for information about the environment. Thirdly the network supplements the computational capacity of the robot. Efficiently designed sensor fusion algorithms can agglomerate intelligence gathered through on-board as well as off board resources to come up with robust decisions. In this work, we developed algorithms for multi robotic traffic control in a world mediated by a network. While single robot navigation mediated by a network is well studied there has been little in the area of multi robotic navigation.
In this work, we present a methodology for coordination of multiple robotic agents moving from one location to another in an environment embedded with a network of agents, placed at strategic locations such as intersections. These intersection agents communicate with robotic agents and also with each other to route robots in a way as to minimize the traffic congestion, thus, resulting in the continuous flow of robots. A robot's path to its destination is computed by the network in terms of the next waypoints to reach. The intersection agents are capable of identifying robots in their proximity based on signal strength. An intersection agent controls the flow of agent traffic around it with the help of the data it collects from the messages received from the robots and other surrounding intersection agents.
The congestion of traffic is minimized using a two layered hierarchical strategy. The objective at the primary level, operating at the intersection, is to reduce congestion at the intersection by reducing the time delay of robots crossing them. The flow of traffic is coordinated by assigning priorities to pathways based on the robot density and its rate of change. In this method the robotic agent hailing from the pathway having the highest agent density and lowest rate of change is allotted the highest priority and the paths of agents with lower priority are attuned to accommodate the paths of the robot with higher priority. The intersection agent allows passage for the robot, that has requested space-time allocation at the intersection, till the point of no conflict in the path of the vehicle through the intersection. The secondary layer maintains coordination between intersection agents and routes traffic such that delay is reduced through effective traffic load balancing. The load balancing of robots over multiple intersections is achieved through coordination between intersection agents by communication of robot densities in different pathways. Routing of a robot from its initial position to its destination in the best available path is done. Individually each robot's path in the environment is distributedly computed by the network as a sequence of waypoints to the goal, each successive waypoint one hop less than the previous. Here, the waypoints are the intersections. Intersection agents coordinate to propagate the robot traffic density information to their neighbors so that at each intersection, the agent guides a robot towards the next waypoint which promises minimum time delay.
In the area of multi-agent traffic control, work has been done based on a first come first served like policy. Our method of coordination at intersections is achieved by considering density and rate of change of density along the incoming pathways to an intersection. Extensive comparisons show the performance gain of the current method over existing ones. Theoretical analysis apart from simulation show the advantages of load balanced traffic flow over uncoordinated allotment of robotic agents to pathways - uniform traffic load is maintained at all intersections and no particular intersection is overwhelmed with robots while others are free of them. Transferring the burden of coordination to the network releases more computational power for the robots to engage in critical assistive activities.
No comments:
Post a Comment