Andreas Gibbesch

Planetary rover mobility simulation on soft and uneven terrain

Rovers on Mars or the Moon for planetary exploration are obtaining increased importance within the spaceflight nations. To achieve full mission success, drivability and mobility in all kinds of complex motion scenarios have to be guaranteed. Here, proper modelling and understanding of the complex wheel–soil interaction, i.e. the terramechanics for flexible and rigid wheels interacting with hard, soft and loose soil, are a major driver for supporting reliable rover design and assisting in testing of the flight model. This paper deals with the terramechanical characteristics for wheel–soil contact dynamics modelling and simulation and its experimental validation on the basis of the future European Mars rover mission ExoMars. The physical contact models are integrated by a multi-body system approach and the performance of the rover mobility will be shown for various driving scenarios on hard and soft soil.

Soft Soil Contact Modeling Technique for Multi-Body System Simulation

In the context of planetary exploration with mobile robots a soil contact model (SCM) for prediction and assessment of locomotion performance in soft uneven terrain has been developed. The SCM approach provides a link between the classical, semi-empirical terramechanics theory of Bekker and the capabilities of multi-body system (MBS) simulation technique for general, full 3D simulations of soil contact dynamics problems. Beyond the computation of contact forces and torques SCM keeps track of the plastic soil deformation during simulation. For this purpose it comprises features such as generation of ruts and displacement of soil material that allow computing typical terramechanical contact phenomena like bulldozing, multi-pass effects and drawbar-pull–slippage relations. Unlike volumetric, Finite Element/Discrete Element Method-like approaches SCM applies exclusively surface oriented algorithms with relatively small complexity constants. Moreover, most of the algorithms are of linear complexity. Therefore, the computational efficiency is quite high and adequate for MBS simulation requirements.

Modellbasierte Regelungsansätze für überaktuierte planetare Rover und robotische Elektromobile

Zusammenfassung Der Beitrag beschreibt modellbasierte Regelungsstrategien für planetare Rover und innovative Elektromobile. Trotz prinzipiell gleichen Aufbaus ihrer Fahrwerke werden doch unterschiedliche Reglerlösungen bevorzugt, deren Bandbreite in diesem Beitrag vorgestellt wird. Beide Systeme sind überaktuierte robotische Fahrzeuge, deren Regler die jeweils unabhängig ansteuerbaren Rad- und Lenkantriebe koordinieren. Im Fall des planetaren Rovers wird ein Fahrzeugmodell mit Rad-Sand-Kontaktdynamik verwendet, das innerhalb eines modellprädiktiven Reglers (MPC) zum Einsatz kommt. Für das wesentlich dynamischere Straßenfahrzeug wird der MPC-Ansatz erweitert. Zur Reduzierung der Steuerungsdimension wird ein invertiertes Modell der Fahrdynamik eingeführt und das Optimierungsproblem auf das Bestimmen von Hilfsgrößen für die verbleibenden Freiheitsgrade reduziert. Abstract The paper introduces model based control strategies for planetary rovers and novel electric vehicles. Despite the same fundamental structure of their chassis different types of controllers are preferred. The bandwidth of potential solutions is shown in the paper. Both systems are overactuated robotic vehicles with controllers for coordinating the individually controlable wheel and steering drives. For planetary rovers a model predictive control (MPC) approach is implemented that considers the specific wheel-soil contact dynamics. The solution for the dynamic road vehicles goes beyond the MPC approach. For minimizing the number of controls an inverted vehicle dynamics model is introduced such that the optimization problem is reduced on computing auxiliary variables only, which are representing the remaining degrees of freedom of the system.

Planetary Rover Mobility: Wheel-Soil Interaction based on Multibody System Approach

Abstract The importance and strenghts of Multibody System (MBS) simulation allows uniquely to investigate a wide range of potential rover configurations and mobility performance on various terrains. Moreover, the importance of the dynamical effects, primarily the suspension mechanisms, can be efficiently taken into account. The main goal of MBS simulation is to reliably predict mobility performance, to reduce the amount of costly prototypes, and to give assistance in field experiments. A great advantage is the integration of the MBS simulation and the very complex tyre-soil interaction into the vehicles conceptual design process. This ranges from kinematic investigations for gradeability, maximum step crossing and side slope driving towards the investigations of tyre-soil interaction with respect to tyre sinkage and rolling resistance. In this paper the focus lies on the simulation of wheel forces and applied torques, and on longitudinal slip of tyres on hard and soft soil for MER and RCL-E type rovers.