Howto Raytracer: Ray / Plane Intersection Theory

Categories:

In this tutorial I will derive how to calculate the intersection of a ray and a plane.

As already stated in my ray / sphere intersection howto, a ray r(t)r(t) can be represented by a point on the ray ee and the ray's direction dd: r(t)=e+tdr(t)=e + t d. The set RR of all points on the ray is then given by: R={r(t)tR}R = {r(t) \mid t \in \mathbb{R}}

Similarly, a plane PP can be represented by a point on the plane pp and by its normal nn. So how do we characterize the points on a plane? The main insight is that given any two points a,ba,b on the plane, the vector from aa to bb, i.e. bab-a, lies itself inside the plane and is thus by definition of the plane's normal nn perpendicular to it. To check for perpendicularity, we check if the dot product (ba)n(b-a)\cdot n is 00. The set of the plane's points xx is then given by P={(xp)n=0xR3}P = {(x-p) \cdot n = 0 \mid x \in \mathbb{R}^3} plane representation

To find the intersection of the ray and the plane now, we have to find the points that are in both sets. So we check if a point r(t)r(t) on the ray also fullfills the plane equation:

(r(t)p)n=0(e+tdp)n=0en+tdnpn=0t=pnendnt=(pe)ndn \begin{aligned} (r(t) - p) \cdot n = 0 \\ (e + t d - p) \cdot n = 0 \\ e \cdot n + t d \cdot n - p \cdot n = 0 \\ t = \frac{p \cdot n - e \cdot n }{d \cdot n} \\ t = \frac{(p - e) \cdot n }{d \cdot n} \end{aligned}

So we just calculate tt, plug it back in the ray's equation r(t)r(t) and end up with the hit point, unless the denominator dnd \cdot n is 00 in which case there is no intersection. Geometrically this corresponds to the ray and the plane being parallel.

Here's some sample code that implements this collision test:

    public override RayTracer.HitInfo Intersect(Ray ray)
    {
        RayTracer.HitInfo info = new RayTracer.HitInfo();

        Vector3 d = ray.direction;
        float denominator = Vector3.Dot(d, normal);

        if (Mathf.Abs(denominator) < Mathf.Epsilon) return info;      // direction and plane parallel, no intersection

        float t = Vector3.Dot(center - ray.origin, normal) / denominator;
        if (t < 0) return info;    // plane behind ray's origin

        info.time = t;
        info.hitPoint = ray.GetPoint(t);
        info.normal = normal;
        return info;
    }

All in all, ray / plane intersection is rather easy if you 're working with inifinitely long planes. If your planes have a width and height, it gets drastically more complicated using just a normal to represent a plane, because this representation is ambiguous to rotations around the normal. However for planes with finite width/height this orientation is important and you would be better suited representing the plane by the (orthogonal) vectors that span it.