The control of an autonomous excavator during the digging process is a complex task as the interaction of the bucket and the soil creates significant external forces and moments on the excavator’s attachment. Due to the harsh working environment of the excavators, it is not practical to install sensors on the attachment of the excavator to measure parameters. Also, a series of external noises such as vibrations, impacts, dust, etc. are affecting the sensor measurements. In this paper, a real-time simulation method based on the marching cube method is presented which captures the initial condition of the soil and estimates the shape of the soil as well as the amount of the soil banked into the bucket. The marching cube method provides an a simple, yet accurate model for estimating soil shape, which eliminates the need for complex and computationally intensive processes required in previous studies using the voxel simulation method. The depth camera also provides sufficient information to capture the soil shape in the noisy environment of construction sites. Thus, the marching cube method combined with a depth camera can produce an accurate ground shape model for monitoring digging procedures. Initially, the shape of the ground will be captured by a depth camera and translated into the marching cube method to represent the current shape of the ground. These two steps are only executed once before starting the excavation process. Then, the pose of the bucket will be determined by the joint angles to define the bucket traverse. The volume created by the bucket motion and marching cube representation of the ground will be subtracted from each other using the constructive solid geometry approach to update the shape of the ground, as well as the amount of the soil which is currently held in the bucket. The proposed method can estimate the soil-bucket interaction without the need for any extra sensory information. Also, the low computational power required to execute the mentioned process makes it viable to run in real time on decent hardware. The other widely used approaches require installing the external sensors on the attachment of the bucket, which is not applicable or demands high computational powers which can not be calculated in real-time. Thus, the proposed method can address the issues of the previously proposed methods.