TABLE 2.

Characteristics of reinforcement learning based coverage and patrol.

Characteristic	Description
Objective	Maximize area coverage and importance (patrolling)
State Representation	Environment states as grid cells, importance information
Action Space	Movement actions and sensing actions
Reward Function	Rewards for total coverage, visiting new areas, patrolling importance
Observation	Environmental sensing target
Challenges	Sparse rewards, large spaces and inhomogeneous patrolling
Citations	Pham et al. (2018); Faryadi and Mohammadpour Velni. (2021); Theile et al. (2020); Luis et al. (2020); Lu et al. (2022); Kouzehgar et al. (2020); Luis et al. (2021); Ai et al. (2021)