Search arXiv⌕ Search

arXiv subjects

Jared Paquet

Publications and source records attributed to Jared Paquet.

3 recordsLinked to original sources

CoPlanVLM: Coordinated Path Planning for a Robot Team using Vision Language Models

We present CoPlanVLM, which uses a Vision Language Model (VLM) to translate a single natural-language instruction and image into coordinated paths for a team of two robots. While VLMs offer a flexible, natural-language interface for commanding robot teams, their weak spatial reasoning has greatly limited VLM-based path planning. CoPlanVLM improves the VLM's spatial reasoning and path planning abilities through three components: (1) an image overlay marking both free space and obstacles, (2) task classification that selects the most appropriate low-level planner and reasoning prompt for each instruction, and (3) task-specific chain-of-thought reasoning that prompts the VLM to identify key objects, allocate roles, and verify routes are collision-free before committing to a plan. Together, these components enable the team to complete complex coordinated maneuvers requiring both robots and divide large tasks between robots to increase efficiency. CoPlanVLM also supports detailed maneuvers involving specific routes, loops, and coverage of large areas, as well as tasks requiring directional reasoning such as north/south or left/right. We validate CoPlanVLM in both Gazebo simulation and on physical Turtlebot 4 hardware. CoPlanVLM attains an average success rate of 84.4% compared to 50.0% for the closest baseline. On trials where both methods succeed, CoPlanVLM's paths are 27% shorter on average. Project code is available at https://github.com/AICPS/CoPlanVLM.

cs.RO↗

Deep Nonlinear Adaptive Control for Unmanned Aerial Systems Operating under Dynamic Uncertainties

Recent literature in the field of machine learning (ML) control has shown promising theoretical results for a Deep Neural Network (DNN) based Nonlinear Adaptive Controller (DNAC) capable of achieving trajectory tracking for nonlinear systems. Expanding on this work, this paper applies DNAC to the Attitude Control System (ACS) of a quadrotor and shows improvement to attitude control performance under disturbed flying conditions where the model uncertainty is high. Moreover, these results are noteworthy for ML control because they were achieved with no prior training data and an arbitrary system dynamics initialization; simply put, the controller presented in this paper is practically modelless, yet yields the ability to force trajectory tracking for nonlinear systems while rejecting significant undesirable model disturbances learned through a DNN. The combination of ML techniques to learn a system's dynamics and the Lyapunov analysis required to provide stability guarantees leads to a controller with applications in safety-critical systems that may undergo uncertain model changes, as is the case for most aerial systems. Experimental findings are analyzed in the final section of this paper, and DNAC is shown to outperform the trajectory tracking capabilities of PID, MRAC, and the recently developed Deep Model Reference Adaptive Control (DMRAC) schemes.

eess.SY↗

Learning Flight Control Systems from Human Demonstrations and Real-Time Uncertainty-Informed Interventions

This paper describes a methodology for learning flight control systems from human demonstrations and interventions while considering the estimated uncertainty in the learned models. The proposed approach uses human demonstrations to train an initial model via imitation learning and then iteratively, improve its performance by using real-time human interventions. The aim of the interventions is to correct undesired behaviors and adapt the model to changes in the task dynamics. The learned model uncertainty is estimated in real-time via Monte Carlo Dropout and the human supervisor is cued for intervention via an audiovisual signal when this uncertainty exceeds a predefined threshold. This proposed approach is validated in an autonomous quadrotor landing task on both fixed and moving platforms. It is shown that with this algorithm, a human can rapidly teach a flight task to an unmanned aerial vehicle via demonstrating expert trajectories and then adapt the learned model by intervening when the learned controller performs any undesired maneuver, the task changes, and/or the model uncertainty exceeds a threshold

cs.RO↗