Unconventional Visual Sensors for Autonomous Vehicles

Abstract

Autonomous vehicles rely on perception systems to understand their surroundings for further navigation missions. Cameras are essential for perception systems due to the advantages of object detection and recognition provided by modern computer vision algorithms, comparing to other sensors, such as LiDARs and radars. However, limited by its inherent imaging principle, a standard RGB camera may perform poorly in a variety of adverse scenarios, including but not limited to: low illumination, high contrast, bad weather such as fog/rain/snow, etc. Meanwhile, estimating the 3D information from the 2D image detection is generally more difficult when compared to LiDARs or radars. Several new sensing technologies have emerged in recent years to address the limitations of conventional RGB cameras. In this paper, we review the principles of four novel image sensors: infrared cameras, range-gated cameras, polarization cameras, and event cameras. Their comparative advantages, existing or potential applications, and corresponding data processing algorithms are all presented in a systematic manner. We expect that this study will assist practitioners in the autonomous driving society with new perspectives and insights.

Control Architecture for Connected Vehicle Platoons: From Sensor Data to Controller Design Using Vehicle-to-Everything Communication

“…By finding a control law (7) so that the cost function (8) is minimised, the states of the platoon are led to follow the dynamics of the states controlled by the DMPC strategy. This means that the behaviour of the platoon (controlled with (7)) is close to the behaviour of the platoon controlled by the DMPC algorithm. Moreover, using a random large set of references, the platoon is tested in multiple operating points, ensuring that no bias from a particular case influences the calculus of matrix K.…”

“…Cameras may be mono, stereo, or full surround, and placed in areas such as dashboards and windshields. These types of cameras used for autonomous vehicles are described in detail in [7]. Depending on the quality of the lens, the maximum working distance of the camera is around 250 m. The main advantages of a camera are related to the accuracy of the colour distribution, the contour of the surroundings, and the texture [23].…”

“…To perform a desired common task for the entire platoon, each vehicle needs to exchange the relevant local measured data with the other participants. Usually, a vehicle platoon task is to travel with a velocity imposed by the leader vehicle of the platoon, while maintaining a desired safe distance between the follower vehicles [7]. One of the most promising functionalities for CAVs is cooperative adaptive cruise control (CACC).…”

Comprehensive Assessment of Artificial Intelligence Tools for Driver Monitoring and Analyzing Safety Critical Events in Vehicles

“…This approach is used to detect certain driver states, such as fatigue or stress; however, drivers often consider the attached sensors to be intrusive. Lastly, the behavioral approach captures driver-facing video recordings to manually or algorithmically assess specific driver behaviors [ 13 , 14 , 15 ].…”