Intelligent houses and buildings, autonomous automobiles, drones, robots, and other items that are successfully incorporated into daily life are examples of autonomous systems and the Internet of Things (IoT) that have advanced as research areas. Secured data transfer in untrusted cloud applications has been one of the most significant requirements in the cloud in recent times. In order to safeguard user data from unauthorised users, encrypted data is stored on cloud servers. Existing techniques offer either security or efficiency for data transformation. They fail to retain complete security while undergoing significant changes. This research proposes novel technique in multipath routing based energy optimization of autonomous networks. The main goal of this research is to enhance the secure data transmission in cloud computing with network energy optimization. The secure data transmission is carried out using multi-authentication attribute based encryption with multipath routing protocol. Then the network energy has been optimized using multi-objective fuzzy based reinforcement learning. The experimental analysis has been carried out based on secure data transmission and energy optimization of the network. The parameters analysed in terms of scalability of 79%, QoS of 75%, encryption time of 42%, latency of 96%, energy efficiency of 98%, end-end delay of 45%.