Real-time measurement and estimation of soil-tool interaction dynamics are key tools for the development of automated excavation systems. Automating a complex operation such as excavation is a challenging task that strongly hinges on a good estimate of soil properties as well as the interaction forces between the excavator tool and the environment. This paper presents a method for identifying all the unknown soil parameters (density, tool-soil friction angle, soil-soil friction angle, and cohesion) required for predicting the interaction forces in real time using a novel hybrid soil model. The hybrid model consists of the Mohr-Coulomb (M-C) soil model and the Chen and Liu upper bound (CLUB) soil model. A switching mode is utilized to select the most appropriate soil model to compute the failure forces that depend on the position of the excavator bucket. The Newton-Raphson method (NRM) is adapted and is used to identify the soil parameters by minimizing the error between the measured forces and the forces computed by the hybrid soil model. The experimental results presented demonstrate that the proposed estimation scheme is accurate when compared to measured soil parameters. The Newton-Raphson estimation method is applied iteratively to efficiently and robustly identify the parameters. The conducted comparative study shows that the proposed method is a promising way for the on-line soil-tool interaction force identification of an automated excavator in dynamic and potentially hazardous environments, outperforming other estimation methods including the least squares method (LSM).