What problem does it solve? Tuning the Nav2 MPPI controller requires balancing nine interacting critic plugins whose weights determine whether a robot follows paths, avoids obstacles, and reaches goals correctly. Without a reference, misconfigured weights cause corner cutting, corridor refusal, goal orbiting, and unwanted reversing. ## Core Features & Use Cases - Per-Critic Parameter Reference: Documents every parameter, default, and behavior for ConstraintCritic, CostCritic, GoalCritic, GoalAngleCritic, PathAlignCritic, PathFollowCritic, PathAngleCritic, PreferForwardCritic, TwirlingCritic, and VelocityDeadbandCritic. - Critic Interaction Analysis: Explains known conflicts such as PathAlignCritic versus GoalCritic near goals and CostCritic versus PathAlignCritic in narrow passages, with concrete resolutions. - Ready-Made Weight Profiles: Provides YAML configurations for open indoor, narrow hallway, and cluttered room environments plus a step-by-step tuning methodology. - Use Case: A robot refuses to enter a 0.9m corridor. Use this Skill to identify that CostCritic weight is too high relative to PathAlignCritic and apply the narrow hallway profile. ## Quick Start Ask the AI to tune the MPPI critics so the robot tracks the global path tightly through narrow hallways without clipping obstacles.
