In order to further expand the flight envelope of quadrotors under actuator failures, we design a nonlinear sensor-based fault-tolerant controller to stabilize a quadrotor with failure of two opposing rotors in the high-speed flight condition (>8 m/s). The incremental nonlinear dynamic inversion approach which excels in handling model uncertainties is adopted to compensate for the significant unknown aerodynamic effects. The internal dynamics of such an underactuated system have been analyzed, and subsequently stabilized by redefining the control output. The proposed method can be generalized to control a quadrotor under single-rotor-failure and nominal conditions. For validation, flight tests have been carried out in a large-scale open jet wind tunnel. The position of a damaged quadrotor can be controlled in the presence of significant wind disturbances. A linear quadratic regulator approach from the literature has been compared to demonstrate the advantages of the proposed nonlinear method in the windy and high-speed flight condition.

Self-learning approaches, such as reinforcement learning, offer new possibilities for autonomous control of uncertain or time-varying systems. However, exploring an unknown environment under limited prediction capabilities is a challenge for a learning agent. If the environment is dangerous, free exploration can result in physical damage or in an otherwise unacceptable behavior. With respect to existing methods, the main contribution of this paper is the definition of a new approach that does not require global safety functions, nor specific formulations of the dynamics or of the environment, but relies on interval estimation of the dynamics of the agent during the exploration phase, assuming a limited capability of the agent to perceive the presence of incoming fatal states. Two algorithms are presented with this approach. The first is the Safety Handling Exploration with Risk Perception Algorithm (SHERPA), which provides safety by individuating temporary safety functions, called backups. SHERPA is shown in a simulated, simplified quadrotor task, for which dangerous states are avoided. The second algorithm, denominated OptiSHERPA, can safely handle more dynamically complex systems for which SHERPA is not sufficient through the use of safety metrics. An application of OptiSHERPA is simulated on an aircraft altitude control task.

A control strategy is proposed to deal with the fundamental gain selection problem of optical flow landings. It involves detecting the height by means of an oscillating movement and setting the control gains accordingly at the start of a landing. Then, during descent, the gains are reduced exponentially, with mechanisms in place to ensure high-performance landings. Real-world experiments with a quadrotor demonstrate successful landings in both indoor and outdoor environments.

This work extends the drag-force enhanced quadrotor model by denoting the free stream air velocity as the difference between the ground speed and the wind speed. It is demonstrated that a relatively simple nonlinear observer is capable of estimating the local wind components, provided accelerometer and GPS-velocity measurements are available. We perform a wind tunnel experiment at various wind speeds using a quadrotor vehicle with a low-cost Inertial Measurement Unit (IMU) and a motion tracking system to provide accurate ground speed measurements. It is shown that the onboard Extended Kalman Filter (EKF) accurately estimates the wind components.

Micro Aerial Vehicles (MAVs) are limited in their operation outdoors near obstacles by their ability to withstand wind gusts. Currently widespread position control methods such as Proportional Integral Derivative control do not perform well under the influence of gusts. Incremental Nonlinear Dynamic Inversion (INDI) is a sensor-based control technique that can control nonlinear systems subject to disturbances. This method was developed for the attitude control of MAVs, but in this paper we generalize this method to the outer loop control of MAVs under gust loads. Significant improvements over a traditional Proportional Integral Derivative (PID) controller are demonstrated in an experiment where the drone flies in and out of a fan's wake. The control method does not rely on frequent position updates, so it is ready to be applied outside with standard GPS modules.

The problem of output optimization within a specified input space of neural networks (NNs) with fixed weights is discussed in this paper. The problem is (highly) nonlinear when nonlinear activation functions are used. This global optimization problem is encountered in the reinforcement learning (RL) community. Interval analysis is applied to guarantee that all solutions are found to any degree of accuracy with guaranteed bounds. The major drawbacks of interval analysis, i.e., dependency effect and high-computational load, are both present for the problem of NN output optimization. Taylor models (TMs) are introduced to reduce these drawbacks. They have excellent convergence properties for small intervals. However, the dependency effect still remains and is even made worse when evaluating large input domains. As an alternative to TMs, a different form of polynomial inclusion functions, called the polynomial set (PS) method, is introduced. This new method has the property that the bounds on the network output are tighter or at least equal to those obtained through standard interval arithmetic (IA). Experiments show that the PS method outperforms the other methods for the NN output optimization problem.

This paper presents the results of the design and evaluation of a context based intent inference system for highway lane changes. Traditionally, intent inference for this task is based on information from the subject vehicle only. Because the context, e.g. traffic situation, plays an important role in the driver's situational awareness, a study was performed to investigate what results can be achieved by including this context information in the intent inference system. It was observed that traditional state based systems have a 90% accuracy and reach their maximum performance much earlier than a simple lateral threshold. The addition of various context information variables to the state based system decreased performance, caused by the models having problems capturing the complexity of the context data.