The recent advancement of the autonomous vehicle has raised the need for reliable environmental perception. This is evident, as an autonomous vehicle has to perceive and interpret its local environment in order to execute reactive and predictive control action. Object Tracking is an integral part of vehicle perception, as it enables the vehicle to estimate surrounding objects trajectories to achieve dynamic motion planning. The 3D LIDAR has been widely used in object tracking research since the mechanically compact sensor provides rich, far-reaching and real-time data of spatial information around the vehicle. On the other hand, the development of autonomous driving is heading toward its use in the urban-driving situation. In an urban situation, a robust detection and tracking algorithm is required due to increasing number of Vulnerable Road User (e.g. pedestrian and cyclist), heterogeneous terrain, inherent measurement uncertainties and limited sensor reach. This thesis presents an integrated framework of multi-target object detection and tracking using 3D LIDAR geared toward urban use. The framework combines occlusion-aware detection methods, probabilistic adaptive filtering and computationally efficient heuristics logic-based filtering to handle uncertainties arising from sensing limitation of 3D LIDAR and complexity of the target object movement. The implemented framework takes a raw 3D LIDAR data as input to perform multi-target object detection while simultaneously maintaining track of the detected objects' kinematic states and dimension in robust, causal, and real-time manner. Robust detection is enabled by slope-based ground removal and L-shape fitting to reliably enclose bounding box to objects of interest in the presence of sensor occlusion. The tracker utilises three combined Bayesian filters (IMM-UK-JPDAF) which simultaneously tackle association uncertainties, motion uncertainties and estimate non-linear stochastic motion model in real-time. Logic-based rule filters are also designed to augment the rest of detection and tracking based on the understanding of LIDAR sensor limitation and occlusion characteristic. The evaluation results using real-world pre-recorded 3D LIDAR data show the proposed framework can achieve promising real-time tracking performance in urban situations. Diverse datasets are deliberately chosen to evaluate if the MOT system is capable of working in a varying driving scenario. The benchmark results highlight that the designed and implemented MOT system is performing on par with the state-of-art works and yield satisfying accuracy and precision in most given urban settings.

This thesis presents a visual route following method that minimizes memory consumption to the point that even Micro Aerial Vehicles (MAV) equipped with only a simple microcontroller can traverse distances of a few hundred meters. Existing Simultaneous Localization and Mapping (SLAM) algorithms are too complex for use on a microcontroller. Instead, the route is modeled by a sequence of snapshots that can be followed back using a combination of visual homing and odometry. Three visual homing methods are evaluated to find and compare their memory efficiency. Of these methods, Fourier-based homing performed best: it still succeeds when snapshots are compressed to less than twenty bytes. Visual homing only works from a small region surrounding the snapshot, therefore odometry is used to travel longer distances between snapshots. The proposed route following technique is tested in simulation and on a Parrot AR.Drone 2.0. The drone can successfully follow long routes with a map that consumes only 17.5 bytes per meter.

Quadrotors  are unmanned aerial vehicles (UAVs) which are controlled by changing the angular speed of the four rotors. In the recent years, they have received a large  attention from the research community thanks to their  vertical take-off and landing capability, small dimension, payload, flight endurance and low price  which it has increased the interest in making them completely autonomous. In this thesis two different problems related to the autonomous quadrotor navigation field have been studied. The first one is how it is possible to drive the  quadrotor from its current pose (position and orientation) to the  desired one assuming that the quadrotor's full states measurements (position, orientation,velocity and acceleration) are available. The second problem deals with the use of visual informations either acquired by the front or bottom camera to place the quadrotor at a desired distance from a chosen static or moving object keeping the  latter centered in the acquired camera image without knowing  neither the quadrotor's pose nor the object one. To solve the first problem  a navigation controller framework able to communicate with both parrot AR. Drone 2.0 and Pixhawk autopilot has been  designed. A cascade control design made up by  seven 2DOF PID controllers divided into five different modules (horizontal position controller, vertical position controller, horizontal speed controller, vertical speed controller, yaw controller) has been developed to ensure that the navigation controller framework  is able to simultaneously track the desired yaw angle together with either the desired velocities or the positions. To solve the second problem  a vision based planner made up by three different modules (perception, image state estimator, image based visual servo) has been developed. Planar ArUco markers have been used to avoid the need of designing different specific objects' detectors. Their corners have been extracted by the ArUco detector located inside the perception module. Inside the image state estimator module a Kalman filter with velocity constant model has been designed for both estimating the corners when a detection is suddenly not available and providing to the image based visual servo module feedbacks data (estimated corners) at a specific chosen frequency. An image based visual servo algorithm  has been implemented to compute the desired quadrotor's translational velocities that the navigation controller framework has to track to minimize the error between the estimated and the desired corners where the latter are computed as a function of the desired distance between the quadrotor and the visual marker. To validate the navigation controller framework and the vision based planner a mission has been chosen in which firstly the quadrotor autonomously reaches a first predefined desired pose, secondly it  approaches up to a desired distance a static visual marker using the front camera image data, thirdly it reaches a  second predefined desired pose and finally it  autonomously lands using the bottom camera image information either on a static or moving visual marker. The mission has been tested in real flight with Parrot AR. Drone 2.0 and in simulation using Gazebo simulator in combination with PX4 Software-In-The-Loop. Both the navigation controller framework and the vision based planner have been developed using ROS as a software framework to guarantee system communication and all the algorithms have been coded using C++ as a programming language.