ICRA2024 Talk: N-MPC for Deep Neural Network-based Collision Avoidance exploiting Depth Images
M. Jacquet, and K. Alexis, "-MPC for Deep Neural Network-based Collision Avoidance exploiting Depth Images", IEEE ICRA 2024
Abstract: This paper introduces a Nonlinear Model Predictive Control (NMPC) framework exploiting a Deep Neural network for processing onboard-captured depth images for collision avoidance in trajectory-tracking tasks with Unmanned Aerial Vehicles. The network is trained on simulated depth images to output a collision score for queried 3D points within the sensor field of view. Then, this network is translated into an algebraic symbolic equation and included in the NMPC, explicitly constraining predicted positions to be collision-free throughout the receding horizon. The NMPC achieves real time control of a UAVs with a control frequency of 100Hz. The proposed framework is validated through statistical analysis of the collision classifier network, as well as Gazebo simulations and real experiments to assess the resulting capabilities of the NMPC to effectively avoid collisions in cluttered environments. The associated code is released open-source.