Cooperative Detection of Multiple Targets by the Group of Mobile Agents

Abstract

The paper considers the detection of multiple targets by a group of mobile robots that perform under uncertainty. The agents are equipped with sensors with positive and non-negligible probabilities of detecting the targets at different distances. The goal is to define the trajectories of the agents that can lead to the detection of the targets in minimal time. The suggested solution follows the classical Koopman’s approach applied to an occupancy grid, while the decision-making and control schemes are conducted based on information-theoretic criteria. Sensor fusion in each agent and over the agents is implemented using a general Bayesian scheme. The presented procedures follow the expected information gain approach utilizing the “center of view” and the “center of gravity” algorithms. These methods are compared with a simulated learning method. The activity of the procedures is analyzed using numerical simulations.

Detection of Static and Mobile Targets by an Autonomous Agent with Deep Q-Learning Abilities

“…This detection problem follows the Koopman framework of search and detection problems [ 3 ] (see also [ 4 , 8 ]) and continues the line of previously developed heuristic algorithms [ 12 , 13 ].…”

“…In previous research [ 12 , 13 ], a similar search and detection problem is solved heuristically by evaluating the decisions made at each step of the search and detection process. In the first algorithm, the agent follows the maximal Expected Information Gain ( ) over the cells that are reachable in a single step from the agent’s current location; in the second algorithm, the agent moves one step toward the maximal expected information gain over all the cells, which is the Center Of View ( ) of the domain; and in the third algorithm, the agent moves toward the center of the distribution or the Center Of Gravity ( ) with respect to the current probability map.…”

“…In contrast to the recently suggested algorithms [ 12 , 13 ], which directly implement one-step decision making, the presented novel algorithm includes learning processes and can be used both with model-free learning for direct online detection and with model-based learning for offline policy planning and further online applications of this policy. Since both learning processes follow the same steps (with the only difference being in the source of information regarding the targets’ location probabilities), below, we outline the Q-max algorithm with model-based learning.…”

“…In this research, we follow the direction of heuristic algorithms for search and detection with false positive and false negative detection errors [ 7 , 11 , 12 , 13 ] and consider the detection of static and moving targets. In addition, we assume that the agent is equipped with an on-board controller that is powerful enough to process deep Q-learning and train neural networks on relatively large data sets.…”

“…In addition, we assume that the agent is equipped with an on-board controller that is powerful enough to process deep Q-learning and train neural networks on relatively large data sets. Similar to previously obtained solutions [ 12 , 13 ], a data set is represented by an occupancy grid [ 14 , 15 ], and the decision making for the probability maps follows the Bayesian approach [ 7 , 8 ].…”