• N.Ajay Nagendra[
• CH.Grace Milongthan Devayani
• A.Shanmukha Tejaswi
• B.Chaytrika Ratna
• D.Sylu Varshitha

Environmental monitoring and restoration need intelligent systems that can deal with both immediate threats and long-term ecological restoration on a large and ever-changing area. Traditional single-agent and centralized control systems may face difficulties in scalability, area coverage, and reaction time when dealing with complex environments. To overcome these issues, this paper presents a cooperative multi-agent framework that combines environmental monitoring, hazard response, and autonomous reforestation. The proposed system works in a partitioned environment where multiple autonomous agents work together to accomplish region-specific tasks. Each agent uses limited local sensing for detecting hazardous situations, performing localized mitigation tasks, and assisting in recovery operations like drone-assisted seed planting. The proposed framework emphasizes decentral- ized decision-making, simultaneous area coverage, and effective resource management to avoid redundant exploration and operation overlap among multiple agents. For the purpose of realizing a stable and scalable form of cooperation, the learning framework is modeled using Multi- Agent Proximal Policy Optimization (MAPPO), which adopts a centralized training and decentralized execution strategy. In the learning process, a centralized critic can make use of global knowledge to ensure stable learning and model the dependencies between agents, while each agent retains its own separate policy for decentralized execution. Through experimental assessment in simulated settings, it is shown that the proposed method outperforms independent learning and rule-based approaches in terms of group efficiency, energy-awareness, and reforestation success. The findings show that cooperative multi-agent learning is a robust and scalable platform for intelligent environmental surveillance and autonomous reforestation, which helps to promote sustainable ecological restoration and future disaster response.

Multi-Agent Systems (MAS), Environmental Monitoring, Autonomous Reforestation, Multi-Agent Proximal Policy Optimization (MAPPO), Decentralized Decision-Making, Cooperative Reinforcement Learning, Drone-Assisted Seed Planting, Hazard Detection and Mitigation, Ecological Restoration, and Intelligent Environmental Surveillance.

"Analysis of Multiagent Deep Reinforcement Learning Algorithms in Drone Based Reforestation ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 3, page no.c448-c457, March-2026, Available :http://www.jetir.org/papers/JETIR2603259.pdf

"Analysis of Multiagent Deep Reinforcement Learning Algorithms in Drone Based Reforestation ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 3, page no. ppc448-c457, March-2026, Available at : http://www.jetir.org/papers/JETIR2603259.pdf