// Copyright 2013 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#ifndef EXAMPLES_DEMO_FLOCK_GOOSE_H_
#define EXAMPLES_DEMO_FLOCK_GOOSE_H_
#include <vector>
#include "ppapi/cpp/rect.h"
#include "vector2.h"
// A Goose. Each goose has a location and a velocity. Implements the
// flocking algortihm described here:
// http://processingjs.org/learning/topic/flocking with references to
// http://harry.me/2011/02/17/neat-algorithms---flocking.
class Goose {
public:
// Initialize a Goose at location (0, 0) no velocity.
Goose();
// Initialize a Goose at the given location with the specified velocity.
Goose(const Vector2& location, const Vector2& velocity);
// Run one tick of the simulation. Compute a new acceleration based on the
// flocking algorithm (see Goose.flock()) and update the goose's location
// by integrating acceleration and velocity.
// @param geese The list of all the geese in the flock.
// @param attractors The list of attractors. Geese have affinity for these
// points.
// @param flockBox The geese will stay inside of this box. If the flock_box
// is empty, the geese don't have boundaries.
void SimulationTick(const std::vector<Goose>& geese,
const std::vector<Vector2>& attractors,
const pp::Rect& flock_box);
// Implement the flocking algorithm in five steps:
// 1. Compute the separation component,
// 2. Compute the alignment component,
// 3. Compute the cohesion component.
// 4. Compute the effect of the attractors and blend this in with the
// cohesion component.
// 5. Create a weighted sum of the three components and use this as the
// new acceleration for the goose.
// This is an O(n^2) version of the algorithm. There are ways to speed this
// up using spatial coherence techniques, but this version is much simpler.
// @param geese The list of all the neighbouring geese (in this
// implementation, this is all the geese in the flock).
// @param attractors The list of attractors. Geese have affinity for these
// points.
// @return The acceleration vector for this goose based on the flocking
// algorithm.
Vector2 DesiredVector(const std::vector<Goose>& geese,
const std::vector<Vector2>& attractors);
// Turn the goose towards a target. The amount of turning force is clamped
// to |kMaxTurningForce|.
// @param target Turn the goose towards this target.
// @return A vector representing the new direction of the goose.
Vector2 TurnTowardsTarget(const Vector2& target);
// Accessors for location and velocoity.
Vector2 location() const {
return location_;
}
Vector2 velocity() const {
return velocity_;
}
private:
// Add a neighbouring goose's contribution to the separation mean. Only
// consider geese that have moved inside of this goose's personal space.
// Modifies the separation accumulator |separation| in-place.
// @param distance The distance from this goose to the neighbouring goose.
// @param gooseDirection The direction vector from this goose to the
// neighbour.
// @param separation The accumulated separation from all the neighbouring
// geese.
// @param separationCount The current number of geese that have contributed to
// the separation component so far.
// @return The new count of geese that contribute to the separation component.
// If the goose under consideration does not contribute, this value is the
// same as |separationCount|.
int32_t AccumulateSeparation(double distance,
const Vector2& goose_direction,
Vector2* separation, /* inout */
int32_t separation_count);
// Add a neighbouring goose's contribution to the alignment mean. Alignment
// is the average velocity of the neighbours. Only consider geese that are
// within |kNeighbourRadius|. Modifies the alignment accumulator |alignment|
// in-place.
// @param distance The distance from this goose to the neighbouring goose.
// @param goose The neighbouring goose under consideration.
// @param alignment The accumulated alignment from all the neighbouring geese.
// @param alignCount The current number of geese that have contributed to the
// alignment component so far.
// @return The new count of geese that contribute to the alignment component.
// If the goose under consideration does not contribute, this value is the
// same as |alignCount|.
int32_t AccumulateAlignment(double distance,
const Goose& goose,
Vector2* alignment, /* inout */
int32_t align_count);
// Add a neighbouring goose's contribution to the cohesion mean. Cohesion is
// based on the average location of the neighbours. The goose attempts to
// point to this average location. Only consider geese that are within
// |kNeighbourRadius|. Modifies the cohesion accumulator |cohesion| in-place.
// @param {!number} distance The distance from this goose to the neighbouring
// goose.
// @param {!Goose} goose The neighbouring goose under consideration.
// @param {!goog.math.Vec2} cohesion The accumulated cohesion from all the
// neighbouring geese.
// @param {!number} cohesionCount The current number of geese that have
// contributed to the cohesion component so far.
// @return {!number} The new count of geese that contribute to the cohesion
// component. If the goose under consideration does not contribute, this
// value is the same as |cohesionCount|.
int32_t AccumulateCohesion(double distance,
const Goose& goose,
Vector2* cohesion, /* inout */
int32_t cohesion_count);
Vector2 location_;
Vector2 velocity_;
};
#endif // EXAMPLES_DEMO_FLOCK_GOOSE_H_
Codebase Browser
chromium