Development of a Haptic Pedal for Human-In-the-Loop Applications

Nowadays, Human-Machine Interfaces (or “HMI” in short) and especially haptic interfaces are widely employed in many fields, thanks to the capability of bidirectionally exchanging information between the human being and the environment through algorithms that manage sensors collected data. Such an approach is often indicated as “digital twin”, which identifies a system characterized by a real-world physical entity continuously exchanging information with a virtual model. The present thesis represents a “digital twin” application dealing with the development of a haptic pedal suitable for different purposes, and capable of reproducing the real braking system feedback without the physical implementation of the related hydraulic circuit. To this proposal, the PhD research deals with the development of a mathematical model representing the “digital twin” virtual core, of a sensor-equipped pedal prototype consisting in the “digital twin” real-world core and of a control system development representing the communication layer between the two above-mentioned worlds. Differently from the already existing haptic pedal applications, the proposed solution is not based only on predefined force-displacement characteristic curves, but target feedback force can be also obtained thanks to the integration of a coupled mechanical-hydraulic brake system mathematical model, where the elements are parameterized in order to manage different system layouts and conditions. The flexibility offered by the adopted methodology permits the modification of some model parameters while the simulation is running in order to account for sudden events, such as air injection in brake system, components wear or failures. The developed set of equations is real-time solved thanks to the adoption of Newton-Raphson and Jacobian-Free-Monotonic-Descent (or “JFMD” in short) algorithms, transmitting a feedback to the user based on the applied force at pedal pad. Based on the state of the art outcomes, on the existing regulations analysis and on ergonomics considerations, a pedal prototype consisting in a slider-crank mechanism has been developed. The design phase has been carried out based on numerical simulations and studying the mechanism kinematics, statics and dynamics, privileging the choice of already available in stock components with low delivery time. To further contain the costs, rapid prototyping techniques have been adopted, too. After having implemented the mathematical model and realized the pedal prototype, the focus has been laid on the control architecture, representing the communication layer between real and virtual worlds. Both admittance and impedance control strategies have been explored and implemented, highlighting for both the approaches advantages and disadvantages. A preliminary testing phase has been carried out to tune the control loop parameters. After that, an experimental campaign has been run for evaluating the developed haptic interface feedback returned to the user under different conditions and pedal layouts. Based on the experimental tests outcomes, the pedal prototype design was enhanced implementing already existing commercial solutions and redesigning some components, adopting less deforming materials and more comfortable layouts. The improved design aimed to reduce forces at the constraints while guaranteeing the requirements in terms of angular and linear displacements. A particular attention has been given to the possibility of mounting the new system both in passenger and race configuration, which are characterized by opposite positioning characteristics. The newly designed pedal prototype has been finally realized and it is ready to be tested.