Media
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October 2023, Detroit, USA
Summary: This presentation centers on our article titled "GP-guided MPPI for Efficient Navigation in Complex Unknown Cluttered Environments," accepted for presentation at IROS 2023.
In this study, we present the GP-MPPI, an online learning-based control strategy that integrates MPPI with a local perception model based on Sparse Gaussian Process (SGP). The key idea is to leverage the learning capability of SGP to construct a variance (uncertainty) surface, which enables the robot to learn about the navigable space surrounding it, identify a set of suggested subgoals, and ultimately recommend the optimal subgoal that minimizes a predefined cost function to the local MPPI planner. Afterward, MPPI computes the optimal control sequence that satisfies the robot and collision avoidance constraints.
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October 2022, Kyoto, Japan
Summary: In this presentation, we introduce our article "Autonomous Navigation of AGVs in Unknown Cluttered Environments: log-MPPI Control Strategy," accepted for publication in the IEEE Robotics and Automation Letters (RA-L) and showcased at IROS 2022.
In this study, we proposed a new method called log-MPPI equipped with a more effective trajectory sampling distribution policy which significantly improves the trajectory feasibility in terms of satisfying system constraints. The key idea is to draw the trajectory samples from the normal log-normal (NLN) mixture distribution, rather than from the Gaussian distribution.
International Conference on Control, Automation, Robotics and Vision (ICARCV), December 2020, Shenzhen, China
Summary: This video showcases the presentation of our article titled "Model Predictive Path Integral Control Framework for Partially Observable Navigation: A Quadrotor Case Study" at the 16th ICARCV 2020.
In this study, we introduced an extension to the classical MPPI framework, empowering the robot (specifically, the UAV) to autonomously navigate in 2D or 3D environments while avoiding collisions with obstacles. The key point of our framework lies in equipping MPPI with a 2D or 3D grid that represents the real-world environment, enabling collision-free navigation without introducing additional complexity to MPPI's optimization problem.