unreal.AnimNode_MotionMatching

class unreal.AnimNode_MotionMatching(blend_weight: float = 0.0, internal_time_accumulator: float = 0.0, notify_recency_time_out: float = 0.0)

Bases: AnimNode_BlendStack_Standalone

Anim Node Motion Matching

C++ Source:

  • Plugin: PoseSearch

  • Module: PoseSearch

  • File: AnimNode_MotionMatching.h

Editor Properties: (see get_editor_property/set_editor_property)

  • blend_option (AlphaBlendOption): [Read-Write] How the blend is applied over time to the bones. Common selections are linear, ease in, ease out, and ease in and out.

  • blend_profile (BlendProfile): [Read-Write] Set Blend Profiles (editable in the skeleton) to determine how the blending is distributed among your character’s bones. It could be used to differentiate between upper body and lower body to blend timing.

  • blend_time (float): [Read-Write] Time in seconds to blend out to the new pose. Uses either inertial blending, requiring an Inertialization node after this node, or the internal blend stack, if MaxActiveBlends is greater than zero.

  • blend_weight (float): [Read-Write] Last encountered blendweight for this node

  • database (PoseSearchDatabase): [Read-Write] The database to search. This can be overridden by Anim Node Functions such as “On Become Relevant” and “On Update” via SetDatabaseToSearch/SetDatabasesToSearch.

  • ignore_for_relevancy_test (bool): [Read-Write] If true, “Relevant anim” nodes that look for the highest weighted animation in a state will ignore this node

  • internal_time_accumulator (float): [Read-Write] Accumulated time used to reference the asset in this node

  • max_active_blends (int32): [Read-Write] Number of max active blending animation in the blend stack. If MaxActiveBlends is zero then blend stack is disabled

  • max_blend_in_time_to_override_animation (float): [Read-Write] if the most relevant (recently added) animation is within MaxBlendInTimeToOverrideAnimation, the new requested blend will take its spot otherwise a new blended will be added to the stack

  • notify_recency_time_out (float): [Read-Write] Window of time after firing a notify that any instance of the same notify will be filtered out.

  • play_rate (FloatInterval): [Read-Write] Effective range of play rate that can be applied to the animations to account for discrepancies in estimated velocity between the movement model and the animation.

  • player_depth_blend_in_time_multiplier (float): [Read-Write] AnimPlayers blend in timer will be incremented PlayerDepthBlendInTimeMultiplier times faster on a deeper blend

  • pose_jump_threshold_time (FloatInterval): [Read-Write] Don’t jump to poses of the same segment that are within the interval this many seconds away from the continuing pose.

  • pose_reselect_history (float): [Read-Write] Prevent re-selection of poses that have been selected previously within this much time (in seconds) in the past. This is across all animation segments that have been selected within this time range.

  • reset_on_becoming_relevant (bool): [Read-Write] Reset the motion matching selection state if it has become relevant to the graph after not being updated on previous frames.

  • search_throttle_time (float): [Read-Write] Minimum amount of time to wait between searching for a new pose segment. It allows users to define how often the system searches, default for locomotion is searching every update, but you may only want to search once for other situations, like jump.

  • should_filter_notifies (bool): [Read-Write] Flag that determines if any notifies from originating from an anim player samples should be filtered or not.

  • should_search (bool): [Read-Write] If set to false, the motion matching node will perform a search only if the continuing pose is invalid. This is useful if you want to stagger searches of different nodes for performance reasons

  • should_use_cached_channel_data (bool): [Read-Write] If set to true, the search of multiple databases with different schemas will try to share pose features data calculated during query build the idea is to be able to share as much as possible the continuing pose features vector across different schemas (and potentially improve performances) defaulted to false to preserve behavior backward compatibility

  • store_blended_pose (bool): [Read-Write] if the number of requested blends is higher than MaxActiveBlends, blend stack will blend and accumulate into a stored pose all the overflowing animations. if bStoreBlendedPose is false, the memory to store the pose will be saved, but once reached the MaxActiveBlends, blendstack will start discarding animations, potentially resulting in animation pops

  • use_inertial_blend (bool): [Read-Write]