Guidance and adaptive control for precision langing on Mars

Since the main objective of the first generation of Mars exploration spacecraft (such as Viking, Mars Pathfinder, and Mars Exploration Rover) was to explore large plains on Mars, there was no need to land the spacecraft at precise locations. This is why it was not necessary to control their landing trajectory. This resulted in landing position uncertainties of the order of 100 km, but this was acceptable, considering the objective of these missions. However, with the second generation of Mars exploration spacecraft (scheduled for 2010), the scientific objectives will require the exploration of specific surface features such as ancient lake beds or polar ice. Therefore, high precision landing systems, called pinpoint landing systems, are necessary to reduce the landing error ellipse to the order of 10 km. Aside from helping to satisfy the scientific objectives, pinpoint landing also helps to avoid hazardous terrain and reduce risks of mission failure. As demonstrated by Wolf [67], the most efficient way to increase the landing accuracy is achieved during the atmospheric entry by guiding and controlling the vehicle trajectory in order to eliminate the dispersions caused at entry and accumulated during the hypersonic phase. Based on the so-called Analytical Predictor-Corrector (APC) algorithm [23], the latest developments on entry guidance and control algorithms proposed an analytical solution for thé reference flight-path angle. Two différent objectives are met by this constant flight-path angle guidance and control law: (1) to bring thé velocity of thé vehicle to its specified value at thé specified altitude (required for thé safe parachute deployment) and (2) to reach thé target position in thé longitudinal and lateral planes. However, for thé first objective (final velocity and final altitude), this guidance and control law assumes that thé atmospheric density encountered during thé entry flight is well known. This represents a major weakness of thé APC algorithm applied to Mars entry. This thesis presents an improvement to thé robustness in achieving thé first objective by combining the constant flight-path angle guidance law with a novel adaptive control scheme, derived from the Simple Adaptive Control (SAC) technique [34]. With the SAC technique, three controllers are adapted in order to maintain stability and bring the Cracking error between a reference model output and the plant output to zero. The greatest difficulty in designing the simple adaptive controller is to determine the weighting matrix coefficients of the three adaptive controllers. In fact, selection of the several weighting coefficients is performed by trial and error, which makes this control law difficult to apply and dependent on the designer expertise. With the novel adaptive control algorithm, only one weighting coefficient is needed, thus greatly simplifying the design procedure. Simulation results are shown to demonstrate the accuracy and the robustness of the proposed guidante and adaptive control system