<p>In this paper, a wireless sensor network is considered which is assisted by unmanned aerial vehicles (UAVs) and empowered by intelligent reflecting surfaces (IRS) to enhance the final performance. The scenario is remote monitoring and gathering information about a disaster-stricken area or a complicated geographical area. We propose a new algorithm to maximize the sum throughput of sensor nodes by jointly optimizing the transmission power of sensor nodes, the location of the UAV as well as the IRS phase shift design. Due to the nonconvexity and NP-hard nature of the problem, we utilize the deep deterministic policy gradient (DDPG) method in the proposed algorithm. We also discuss the computational complexity and time consumption of the proposed algorithm. We compare our proposed algorithm with baseline scenarios as well as another reinforcement learning algorithm (i.e., proximal policy optimization). Simulation results validate the efficiency of our proposed algorithm. Also, simulation results show that employing IRS can considerably improve the sum throughput while using less power consumption compared to traditional wireless sensor networks without IRS. Our proposed algorithm can be applied in emergency scenarios and can serve as a platform for further research in wireless sensor networks.</p>