With the recent advancements in self-driving vehicles, it is only a matter of time before autonomous vehicles will be used on public roads. Before this technology is widely adopted, it is vital to ensure that the safety of other road users is considered so as to prevent road traffic-related incidents. These road users include pedestrians, cyclists, motorcyclists, and other vehicle users. Of these road users, pedestrians and cyclists are classed as vulnerable road users (VRUs). For this reason, pedestrian and cyclist detection has received significant attention. Therefore, it is pivotal that other road users, and especially VRUs, meet a level of safety while self-driving vehicles are on public roads. More recently, new machine learning algorithms, more specifically Deep Learning, have been implemented in order to provide unprecedented levels of performance. With such improvements, robotic vehicles can be designed to move closer to becoming fully autonomous, creating safer public roads for all road users.
To address this task, robotic vehicles need to be equipped with sensor networks (i.e., network of inter-connected sensors) to perceive the robot’s immediate surroundings. In this way, the robot is able to determine the safest path to follow with respect to the safety of road users. This provides a high level of safety as well as providing efficiency and comfort, as harsh braking and acceleration will be limited. Machine learning algorithms can be employed to learn from the output of the sensor network so as to detect and predict the future intentions of objects. With the combination of various sensors (e.g., visual, thermal IR, and LIDAR) and using effective machine learning methods, a high safety for road users can be achieved. Certain sensors, such as thermal sensors, have become more accessible because of a decrease in costs, allowing further research to be conducted into sensor fusion.
The aim of this Special Issue is to present the current state-of-the-art in machine learning methods and sensor systems used in robotic vehicles (both urban and non-urban). This Special Issue focuses on the following areas for contribution (but is not limited to them):
Robotic (self-driving) vehicles
Sensor systems in autonomous vehicle
Design of sensor networks
Sensor data processing
Machine learning/deep learning for sensor data
Robotic environmental interactions
Application of sensors for robotics
Sensor data fusion
Robotic (self-driving) vehicle safety
Robotic (self-driving) vehicle efficiency
Dr. Md Nazmul Huda
Dr. Tatiana Kalganova
Prof. Dr. Vasile Palade