A Novel Method for Goal Recognition from 10 m Distance Using Deep Learning in CanSat

In this study, we propose a method for CanSat to recognize and guide a goal using deep learning image classification even 10 m away from the goal, and describe the results of demonstrative evaluation to confirm the effectiveness of the method. We applied deep learning image classification to goal recognition in CanSat for the first time at ARLISS 2019, and succeeded in guiding it almost all the way to the goal in all three races, winning the first place as overall winner. However, the conventional method has a drawback in that the goal recognition rate drops significantly when the CanSat is more than 6–7 m away from the goal, making it difficult to guide the CanSat to the goal when it moves away from the goal because of various factors. To enable goal recognition from a distance of 10 m from the goal, we investigated the number of horizontal regions of interest divisions and the method of vertical shifts during image recognition, and clarified the effective number of divisions and recognition rate using experiments. Although object detection is commonly used to detect the position of an object from an image by deep learning, we confirmed that the proposed method has a higher recognition rate at long distances and a shorter computation time than SSD MobileNet V1. In addition, we participated in the CanSat contest ACTS 2020 to evaluate the effectiveness of the proposed method and achieved the zero-distance goal in all three competitions, demonstrating its effectiveness by winning first place in the comeback category.

Development of Rover with ARLISS Requirements and the Examination of the Rate of Acceleration that Causes Damages During a Rocket Launch

In recent years, CanSats have become a popular choice in simulated satellite contests. Among CanSat contests, the ARLISS project is the one that uses rockets to launch the CanSat into the sky. ARLISS provides rockets to launch CanSats, which reach an altitude of ∼4,000 m and then drop the rover to the ground using a parachute. However, the rovers of several teams cannot withstand the large acceleration applied during the launch, which damage and make them non-operational. The acceleration applied to the rocket during the launch was measured by multiple teams previously; however, because the CanSat is a small-embedded device, an acceleration sensor with a wide measurement range and a high sampling frequency could not be used. In this study, we measure the acceleration applied to the rover from the launch till it drops on the ground using an acceleration sensor with a wider measurement range, and by acquiring data at a higher sampling frequency than before. The acceleration is found to be larger than that in the conventional measurement when the rocket is launched and it drops to the ground. Further, the descriptions of the technical details of the rover structure that can withstand these impacts, perform accurate measurements, and operate without breakage in ARLISS are provided.