IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 10, NO. 2,APRIL 2005 221 Online Soil Parameter Estimation Scheme Based on Newton–Raphson Method for Autonomous Excavation Choo Par Tan, Associate Member, IEEE, Yahya H. Zweiri, Kaspar Althoefer, Member, IEEE, and Lakmal D. Seneviratne, Member, IEEE Abstract—The automation of the excavation process has a huge potential in many industries that require earth removal. The online estimation of soil parameters is an important requirement for de- veloping an impedance controller architecture for automated exca- vation strategies. This paper presents a fast and robust technique for the experimental identification of soil. This technique, based on the Newton–Raphson method, estimates the unknown parame- ters of the soil mechanics equations by minimizing the error be- tween measured failure forces and estimated failure forces. The measured failure forces can be obtained by measuring the forces acting on the bucket during the excavation operation, while the estimated failure forces are obtained by an analytical soil model. The Mohr–Coulomb soil model and the Chen and Liu upper bound soil model are employed in this soil parameter estimation scheme. The proposed estimation method is compared to the parameter space intersection method previously employed for parameter es- timation and has been tested using four different types of soils (Ti- cino, Rained Ticino, Nevada Fine, and Glass Beads). The results show that the proposed technique is in good agreement with the parameter space intersection method, but outperforms the latter in terms of computational execution time. A further disadvantage of the parameter space intersection method is that it relies on the search space to be stored in a tabular form. Hence, if high accu- racy is needed, the discretization step of the tabular data has to be very small, leading to high memory requirements. Further, in con- trast to the method proposed here, the parameter space intersec- tion method suffers from the curse of dimensionality, which leads to exorbitant memory requirements when searching in high-di- mensional parameter spaces. The presented results show that the proposed technique has high robustness to initial condition vari- ations and is very insensitive to perturbations of the input signal. The technique presented in this paper is generic and suitable for a real-time soil parameter estimation scheme. Index Terms—Autonomous excavations, Newton–Raphson method, soil model, soil parameter estimation. I. INTRODUCTION I NDUSTRIES where earth-moving is a predominant oper- ation are the primary users of excavators. Backhoes and front-end loaders are commonly used excavators in agricultural, mining, construction, and military applications. Normally, the Manuscript received May 27, 2003; revised March 8, 2004. This work was supported in part by QinetiQ Ltd. C. P. Tan, K. Althoefer, and L. D. Seneviratne are with the Department of Mechanical Engineering, King’s College London, London, WC2R 2LS, U.K. (e-mail: k.althoefer@kcl.ac.uk). Y. H. Zweiri is with the Department of Mechanical Engineering, King’s Col- lege London, London WC2R 2LS, U.K., currently on leave from the Department of Mechanical Engineering, Mu’tah University, Jordan. Digital Object Identifier 10.1109/TMECH.2005.844706 overall task of an excavator can be divided into a series of actions that are executed consecutively and repetitively by a human operator until a predetermined goal is reached. Due to the repetitive nature of the task, operators commonly suffer from fatigue after long periods of excavation. This can lead to suboptimal performance possibly resulting in reduced effi- ciency and accelerated equipment wear and tear. Also, for the excavator to be used efficiently and effectively, the operator needs to be highly trained and the training is usually costly and time consuming. In remote operations of excavators deployed in hazardous environments, teleoperation is used. However, force feedback is generally not available to the operator; commonly they rely on video feedback only, and thus have difficulties to appropriately control the bucket trajectory during the dig when trying to optimize the bucket load and reduce wear. Obviously, the automation of the excavation process repre- sents a huge potential in many industries that require earth re- moval. It offers many benefits including increased productivity, continuous operation, increased safety levels, and operation in hazardous environments. To develop the desired relationship be- tween the bucket and the soil, an impedance controller can be employed. The soil parameters are estimated online in order to realize the appropriate soil–tool interaction. Based on different soil models, this paper researches an approach for the estimation of soil parameters which can be used in an impedance controller for automated excavation. The soil parameters (soil friction angles and soil density) rep- resent the strength and resistance of the soil. By knowing the strength and soil resistance, the forces acting on an excavation bucket can be predicted. The predicted forces can hence be uti- lized to improve the digging strategy by avoiding overloading that could damage the machine. It can also provide the operator information on the hardness of the soil to plan a better digging strategy. The soil can be classified into different categories based on the soil parameters, and the digging cycle can be optimized taking such soil parameters into consideration. Thus, the effi- ciency and productivity of the excavation can be increased and, the associated cost can be reduced. Although many studies aim to develop fully automated ex- cavators in the long term [2]–[4], [8], [9], current research at- tempts to provide solutions to aid the human operator. A robotic excavation strategy based on a fuzzy behavior formulation for a robotic front-end loader is presented in [5] and [6]. An auto- matic digging-control system for hydraulic excavators was de- veloped by the researchers at Hitachi Construction Machinery 1083-4435/$20.00 © 2005 IEEE