This paper deals with study and tests on an experimental car with antilock-braking system (ABS) and vehicle speed estimation using fuzzy logic. Vehicle dynamics and braking systems is complex and behave strongly non-linear which causes difficulties in developing a classical controller for ABS. Fuzzy logic, however facilitates such system designs and improves tuning abilities.

The design of a symbolic sensor that identifies the condition of the runway surface (dry, wet, icy, etc.) during the braking of a commercial aircraft is also discussed. The purpose of such a sensor is to generate qualitative, real-time information about the runway surface to be integrated into a future aircraft ABS. It can be expected that this information can significantly improve the performance of ABS. For the design of the symbolic sensor different classification techniques based upon fuzzy set theory and neural networks are proposed.

Recently there has been a growing interest in intelligent control techniques for the design of aircraft and road vehicle Antilock Brake Systems (ABS). In particular, rule-based, fuzzy logic controllers have been applied to this problem and successfully tested in simulation. In fact, the use of non-linear, fuzzy control techniques appears to be particularly appropriate for the ABS control problem because of the high non-linearity of the system and the lack of a precise physical model of the friction force between tyre and runway. In addition to that, the controller must operate at an unstable equilibrium point to achieve an optimal braking performance. The most important problem in ABS control design - fuzzy or conventional - is that the optimum adhesion coefficient varies significantly with the surface condition (i.e. dry, wet, icy, etc.) of the runway. Because the latter is unknown, it is extremely difficult to define a controller that guaranties an optimal braking performance for all types of runway conditions.

The underlying control philosophy takes into consideration wheel acceleration as well as wheel slip in order to recognize blocking tendencies. The knowledge of the actual vehicle velocity is necessary to calculate wheel slips. This is done by means of a fuzzy estimator, which weighs the inputs of a longitudinal acceleration sensor and four wheel speed sensors. If lockup tendency is detected, magnetic valves are switched to reduce brake pressure. Performance evaluation is based both on computer simulations and an experimental car. To guarantee real-time ability (one control cycle takes seven milliseconds) and to relieve the electronic control unit (ECU), all fuzzy calculations are made by the fuzzy coprocessor SAE 81C99A. Measurements in the experimental car prove the functionality of this automotive fuzzy hardware system.

Conventional ABS (air craft):

The role of the ABS is to control the wheel speed in order to prevent the wheels from locking and to assure a maximum braking force. This is of major importance when the runway is slippery or very short. Wheel moment of inertia, Rotational wheel speed, Friction coefficient, Wheel Radius Normal force per wheel Friction force, Aircraft speed Figure 1b: are analogous to the above vehicle wheel shown. Forces acting on a braked wheel are shown in fig1b.The ABS commands the brake pressure as a function of the difference between the measured and the reference wheel speed. The latter is calculated from the measured aircraft speed and the desired wheel slips, using equation. At the moment, when the pilot pushes the brake pedal the brake pressure and the wheel slip increase provoking a ground force between tyre and runway. Assuming the case of full braking, the ABS will control the wheel speed to its reference value. To achieve a maximum braking force the reference slip should be chosen close to the optimum slip. However, when the pressure level in the brakes becomes too high, the wheel slip slides beyond the optimum of the adhesion curve and the system tyre/runway becomes instable.

The slope of u(s) being negative, the wheel immediately starts to lock. In this case, the ABS rapidly releases the brake pressure to force the wheel speed back to the stable side of the adhesion curve. In fact, this situation occurs, when either the desired slip s, has been chosen on the instable side of the friction characteristic, or when a sudden change in ground force is encountered (e.g. a transition from a dry to a wet runway surface). The principle problem in ABS design is that the optimum slip and the exact shape of the adhesion characteristic depend on the runway surface and further parameters, which cannot be measured, such as the condition of the tyres or the dynamics of the normal forces. Bearing in mind that the optimum slip value may vary between 3% and 20%, it is clear that the choice of the reference slip value is crucial for a safe and efficient ABS. If it is too small the braking force might become insufficient, if it is too high, wheel lockup occurs.

The design of a symbolic sensor that identifies the condition of the runway surface (dry, wet, icy, etc.) during the braking of a commercial aircraft is also discussed. The purpose of such a sensor is to generate qualitative, real-time information about the runway surface to be integrated into a future aircraft ABS. It can be expected that this information can significantly improve the performance of ABS. For the design of the symbolic sensor different classification techniques based upon fuzzy set theory and neural networks are proposed.

Recently there has been a growing interest in intelligent control techniques for the design of aircraft and road vehicle Antilock Brake Systems (ABS). In particular, rule-based, fuzzy logic controllers have been applied to this problem and successfully tested in simulation. In fact, the use of non-linear, fuzzy control techniques appears to be particularly appropriate for the ABS control problem because of the high non-linearity of the system and the lack of a precise physical model of the friction force between tyre and runway. In addition to that, the controller must operate at an unstable equilibrium point to achieve an optimal braking performance. The most important problem in ABS control design - fuzzy or conventional - is that the optimum adhesion coefficient varies significantly with the surface condition (i.e. dry, wet, icy, etc.) of the runway. Because the latter is unknown, it is extremely difficult to define a controller that guaranties an optimal braking performance for all types of runway conditions.

The underlying control philosophy takes into consideration wheel acceleration as well as wheel slip in order to recognize blocking tendencies. The knowledge of the actual vehicle velocity is necessary to calculate wheel slips. This is done by means of a fuzzy estimator, which weighs the inputs of a longitudinal acceleration sensor and four wheel speed sensors. If lockup tendency is detected, magnetic valves are switched to reduce brake pressure. Performance evaluation is based both on computer simulations and an experimental car. To guarantee real-time ability (one control cycle takes seven milliseconds) and to relieve the electronic control unit (ECU), all fuzzy calculations are made by the fuzzy coprocessor SAE 81C99A. Measurements in the experimental car prove the functionality of this automotive fuzzy hardware system.

Conventional ABS (air craft):

The role of the ABS is to control the wheel speed in order to prevent the wheels from locking and to assure a maximum braking force. This is of major importance when the runway is slippery or very short. Wheel moment of inertia, Rotational wheel speed, Friction coefficient, Wheel Radius Normal force per wheel Friction force, Aircraft speed Figure 1b: are analogous to the above vehicle wheel shown. Forces acting on a braked wheel are shown in fig1b.The ABS commands the brake pressure as a function of the difference between the measured and the reference wheel speed. The latter is calculated from the measured aircraft speed and the desired wheel slips, using equation. At the moment, when the pilot pushes the brake pedal the brake pressure and the wheel slip increase provoking a ground force between tyre and runway. Assuming the case of full braking, the ABS will control the wheel speed to its reference value. To achieve a maximum braking force the reference slip should be chosen close to the optimum slip. However, when the pressure level in the brakes becomes too high, the wheel slip slides beyond the optimum of the adhesion curve and the system tyre/runway becomes instable.

The slope of u(s) being negative, the wheel immediately starts to lock. In this case, the ABS rapidly releases the brake pressure to force the wheel speed back to the stable side of the adhesion curve. In fact, this situation occurs, when either the desired slip s, has been chosen on the instable side of the friction characteristic, or when a sudden change in ground force is encountered (e.g. a transition from a dry to a wet runway surface). The principle problem in ABS design is that the optimum slip and the exact shape of the adhesion characteristic depend on the runway surface and further parameters, which cannot be measured, such as the condition of the tyres or the dynamics of the normal forces. Bearing in mind that the optimum slip value may vary between 3% and 20%, it is clear that the choice of the reference slip value is crucial for a safe and efficient ABS. If it is too small the braking force might become insufficient, if it is too high, wheel lockup occurs.