This paper presents a probabilistic approach for coordinating multiple robots in exploring an unknown environment. The goal is to minimize the overall exploration time by assigning different target points to individual robots so they explore different regions. The approach considers both the cost of reaching a target point and the utility of the target point, which is defined by the unexplored area that can be covered by the robot's sensors. Once a target point is assigned to a robot, the utility of unexplored areas visible from that point is reduced for other robots. This ensures that each robot explores different parts of the environment. The technique uses occupancy grid maps to represent the environment and integrates maps from different robots to create a global map. The robots then select target points based on a trade-off between the cost of reaching the target and its utility. The cost is calculated using a value iteration algorithm, while the utility is estimated based on the expected area that can be covered by the robot's sensors. The approach also considers the visibility range of each robot to avoid conflicts in target selection. The method has been tested in real-world experiments and simulations. The results show that the coordination technique significantly reduces exploration time compared to previous approaches. In experiments with two robots, coordinated exploration reduced the time needed to explore an environment from 49 seconds to 35 seconds. Simulation experiments also demonstrated that coordinated robots were faster than uncoordinated ones, with two coordinated robots performing as well as three uncoordinated robots. The approach differs from previous techniques by explicitly coordinating robot actions and considering both the cost and utility of target points. It also uses a more accurate distance estimation based on the current map rather than straight-line distance. The method is robust and efficient, and further improvements could include more sophisticated strategies for exploration and handling situations where robots do not know their relative positions.This paper presents a probabilistic approach for coordinating multiple robots in exploring an unknown environment. The goal is to minimize the overall exploration time by assigning different target points to individual robots so they explore different regions. The approach considers both the cost of reaching a target point and the utility of the target point, which is defined by the unexplored area that can be covered by the robot's sensors. Once a target point is assigned to a robot, the utility of unexplored areas visible from that point is reduced for other robots. This ensures that each robot explores different parts of the environment. The technique uses occupancy grid maps to represent the environment and integrates maps from different robots to create a global map. The robots then select target points based on a trade-off between the cost of reaching the target and its utility. The cost is calculated using a value iteration algorithm, while the utility is estimated based on the expected area that can be covered by the robot's sensors. The approach also considers the visibility range of each robot to avoid conflicts in target selection. The method has been tested in real-world experiments and simulations. The results show that the coordination technique significantly reduces exploration time compared to previous approaches. In experiments with two robots, coordinated exploration reduced the time needed to explore an environment from 49 seconds to 35 seconds. Simulation experiments also demonstrated that coordinated robots were faster than uncoordinated ones, with two coordinated robots performing as well as three uncoordinated robots. The approach differs from previous techniques by explicitly coordinating robot actions and considering both the cost and utility of target points. It also uses a more accurate distance estimation based on the current map rather than straight-line distance. The method is robust and efficient, and further improvements could include more sophisticated strategies for exploration and handling situations where robots do not know their relative positions.