Perceiving and analyzing the environment around us is a major challenge in many promising industrial sectors. In this context, artificial intelligence (AI) algorithms have undoubtedly demonstrated their effectiveness for tasks related to vision, with various sensors (camera, lidar, etc.). Today, there is a growing interest in the use of AI for radar sensor data (radio detection and ranging). Radar is indeed a sensor that stands out due to the nature of its data, its operability (low light, bad weather, etc.), and its cost. However, they produce sparse data with low spatial resolution, making them difficult to exploit with traditional algorithms. Recently, artificial neural networks based on a graph representation of data (Graph Neural Networks - GNN) have shown good accuracy on sparse and noisy sensor data [1]. Consequently, the use of GNN for radar data exploitation seems very promising [2]. The range of applications is wide, including intelligent vehicles (cabin monitoring), medical devices (vital sign measurements), gesture detection [3], or surveillance devices (fall detection).

In a rapidly evolving context with strong industrial interest, the intern will implement and propose innovative methods for processing data from a radar sensor. They will rely on AI algorithms based on GNNs currently being developed within the laboratory. The student will be integrated into a dynamic multidisciplinary team and will benefit from upskilling in artificial neural networks.

#CeaList

Desired profile: Student in the final year of engineering school or Master 2

Desired skills: A strong motivation to learn and contribute to research in artificial intelligence. In-depth knowledge of computer science and programming languages (Python). Knowledge of artificial intelligence and experience with artificial neural networks (libraries Pytorch or Tensorflow) are a plus. The recruitment interview may refer to the three publications cited.