In this post I reviewed the article “Autonomous landing of an UAV with a ground-based actuated infrared stereo vision system”. In this paper the authors suggest using infrared cameras to aid in UAV landing. The infrared camera is chosen as the exteroceptive sensor for two main reasons: first, it can be used in day and night operations under all-weather conditions; second, it can be used in the environments, where GPS signal is restricted or unavailable due to terrain, obstacles or intentional signal jamming. Another advantage of using infrared stereo cameras according to the article is the system cost and complexity reduction.

Navigation of the UAV is defined as the process of data collection, data analysis, monitoring of the UAV vehicle status and its surrounding, with the goal of successful and safe mission completion. From this statement, we can see that information from proprioceptive (Internal status) and exteroceptive (outside, environmental) sensors combined together is important for success of the mission.

There are four core functions in a navigation system, they are Sensing, State Estimation, Perception, and Situational Awareness. With regard to these four functions, different types of sensors such as the Global Navigation Satellite System (GNSS), laser range scanners (LRFs), monocular cameras, and stereo cameras are been used. In this article, the authors focus on the infrared sensor as a stand along landing system or as an additional system used together with above mentioned sensors.

Considering that the landing phrase of UAV operations is one of the most complex and dangerous segments of flight, precise and timely sensory information is important for safely executing the landing maneuver. The authors point out that it is important to build in extra redundancy into the system, which is responsible for the landing phase. The main idea is to track the UAV during the landing phase and calculate the relative position between the UAV and its landing sight, based on the infrared vision system. The diagram of the system architecture including on-board sensor equipment, ground station and stereo camera sensor architecture is represented in Figure 1.

Figure 1. Architecture of the system.

The authors built and tested a calibrated binocular infrared landing system to estimate relative position between UAV and landing site. The stereo vision system is built on two infrared cameras, with model IRT301, which are produced by IRay Technology. The equipment architecture of the infrared vision landing system is presented in Figure 2.

Figure 2. Ground stereo vision landing system.

This experimental sensing system was tested using field trials, with a quadrotor and a fixed-wing aircraft. This landing system features large field of view buy using pan tilt unit (PTU) represented in the Figure 3.

 

Figure 3. Infrared video system with point-tilt unit.

Testing of the infrared based landing system had positive results. Some of the problems which were experienced during testing relate to low accuracy of fixed-wing UAV touch-down points. However, the use of the proposed system greatly increased situational awareness, aided in navigation, and in 3D position estimation methods.

Reference:

Kong, W., Zang, D., Wang, X., Xiang, Z., & Zhang, J. (n.d.). Autonomous landing of an UAV with a ground-based actuated infrared stereo vision system. Retrieved from https://tams.informatik.uni-hamburg.de/people/kong/publications/130322.IROS2013.pdf