First year: (I) Vectors, dot and cross product Vectors The actual definition of vectors and vector spaces is quite theoretical, but as far as first year maths is concerned, a vector u is something that looks like that: x1 x u 2 . ... x n x1 ,..., xn are the coordinates of the vector u . x1 ,..., xn are real or complex numbers, but for the moment we suppose that they are real. In practice you will only have to deal with vectors having only two or three coordinates, but we will stick to the notation above in order to avoid saying the same thing twice for n 2 and n 3 . Two-dimensional vectors belong to R2 and three-dimensional vectors belong to R3. Let’s see how two-dimensional vectors work: 2 3 1 2 Let u1 , u 2 , u3 and u 4 . See if you can spot those 3 2 1 1 vectors on the graph below: Three-dimensional vectors are a bit trickier to plot. 2 3 1 2 Let v1 3 , v2 2 , v3 1 and v4 1 . 3 3 4 4 The extra dimension makes them harder to spot: A set S {u1 ,..., uk } of vectors is said to be linearly independent if 1u1 ... k uk 0 implies 1 ... n 0 . 1 0 2 For example, let u1 , u 2 and u3 . Then 0 1 3 0 2u1 3u 2 u3 0 so that S {u1 , u 2 , u3 } is not linearly independent. On the other hand, the set {u1 , u2 } is linearly independent since 0 1u1 2u 2 1 implies 1 2 0 . 2 0 The fact that a set is not linearly independent means that some vectors in the set can be expressed as linear combinations of other vectors in the set. For example S (as above) is not linearly independent and u3 2u1 3u 2 . A basis for Rn ( n 2 or n 3 ) is a set of n linearly independent vectors. For example {u1 , u2 } (as above) is a basis for R2 since {u1 , u2 } consists of two linearly independent vectors. If a set S of vectors is a basis for Rn, then every vector in Rn can be written as a linear combination of vectors in S . (II) Dot product The dot product is also sometimes called the scalar product. x1 y1 x2 y Let u and v 2 . ... ... x y n n The dot product of u and v is u v x1 y1 x2 y2 ... xn yn . 1 4 For example, say u 2 and v 5 . Then u v 1 4 2 5 3 6 32. 3 6 It is important to keep in mind that the dot product of two vectors is a number and not a vector. Two vectors u and v are said to be orthogonal if u v 0 . 2 1 2 . Then u v 2 1 3 0. For example, let u and v 3 3 2 / 3 If you plot u and v , it becomes rather obvious that they are orthogonal: Given a linearly independent set of vectors {u1 , u 2 ,..., u n } in Rn, it is possible to turn it into an orthogonal set (i.e. a set where the vectors are orthogonal to each other) using the Gram-Schmidt process. Gram-Schmidt process: Take the first vector in your set, in this case u1 and call it v1 (just to initiate the process). Then set e1 v1 v1 v1 e1 is the first vector in your orthogonal set. Now set v2 u2 (u2 v1 )v1 and e2 v2 . v2 v2 e2 is the second vector in your orthogonal set. For the third vector we proceed similarly: v3 u3 (u3 v1 )v1 (u3 v2 )v2 and e3 v3 . v3 v3 Etc… If you like formulae, the following one gives you the k th vector of the orthogonal set vk uk (uk v1 )v1 (uk v2 )v2 ... (uk vk 1 )vk 1 and ek vk vk vk 1 2 For example, let’s take u1 , and u 2 . 0 3 Set v1 u1 . Since v1 v1 1 , we can also set e1 u1. Now, set v2 u2 (u2 v1 )v1 . We have u2 v1 1 2 0 3 2 so that 0 2 1 0 v v2 2 and e2 2 . 32 1 3 0 3 . (III) Cross Product When it comes to the cross product, we have to take n 3 , i.e. we consider three-dimensional vectors. 1 0 0 Let i 0 , j 1 and k 0 . {i, j, k} is a basis, so every vector in R3 can be 0 0 1 described as a linear combination of the vectors in the basis. Namely, if x1 u x2 then u x1 i x2 j x3 k . x 3 The cross product of two vectors u and v is a vector that is orthogonal to both u and v and, more precisely, the cross product of u and v is given by the following formula: x1 y1 x u x2 , v y 2 u v i 2 y2 x y 3 3 x3 y3 j x1 x3 y1 y3 k x1 x2 y1 y2 . There is a trick to remember the formula for the cross product: i u v x1 y1 j x2 y2 k x3 y3 you can treat the right-hand side like the determinant of a 3 3 matrix and expand with respect to the first row. Of course, this is just a trick, not a proper formula. For example, take i j k 1 1 0 u 0 and v 1 . Then u v 1 0 1 i j k 1. 1 1 1 0 1 1 Check that u (u v) 0 and that v (u v) 0 , so that u v is indeed orthogonal to both u and v . With three vectors u, v and w , you can form w (u v ) . This dot-cross product is sometimes called the triple product. z1 x1 y1 z1 If u x2 , v y 2 and w z 2 , then w (u v) x1 x y z y1 3 3 3 z2 x2 z3 x3 . y2 y3