In this study, in order to increase the energy efficiency in autonomous excavators, which are great solutions for the high labor costs and harsh and hazardous environmental conditions of the construction industry, two approaches have been proposed. First, a new and unique design with two parallel arm and bucket actuators has been proposed for the electric excavator manipulator. In the conventional design for excavators, the three actuators of the boom, the arm, and the bucket are in series. Therefore, they cannot share the external load between them. However, in the proposed new design, the arm and the bucket actuators are in parallel, which helps them to share the load between themselves. This approach helps the excavator to overcome the higher external load with the same actuators. Additionally, in this design, to reduce the energy consumption during the idling time of the excavator, the hydraulic actuators have been replaced with electric linear actuators. Since these actuators have low back drivability, they can handle relatively high external forces without spending energy while they are not in motion. Secondly, a PSO-based path generation algorithm has been developed for autonomous excavation to minimize energy consumption, while it's trying to generate a path as close as possible to the desired path, and avoid collision with undesired obstacles. In the end, two scenarios have been considered to test the performance of the algorithm to save energy. Since in the PSO algorithm, by changing the gains in the cost function, it's possible to change the priorities of the elements to minimize, simulations have been performed to compare results between scenarios with and without considering energy-saving. The simulations show that the proposed algorithm considering the energy-saving decreases the energy consumption in each cycle of digging by 18.51%.