Once I position the camera I then needed to change the camera view from perspective to orthographic. Orthographic view simply squashes the map into two-dimensional form and gets rid of the depth of the map which is visible in perspective view.
I now needed to change the depth value of the camera to 1 and the main game camera to 0. 0 is issued to the camera which you want to be used as the main camera and then other cameras in the screen are numbered correspondingly.
Now I needed to go the game window and re-size the scale of the camera's view and position it in the bottom left hand corner of the screen. This is done simply by messing around with the X, Y, Height and Width values until I was happy.
This would be fine however I do not need to add such a high detailed mini map. All I want is to display the terrain, road and agents in the level and the agents them selves are very small to see.
To solve this I needed to duplicate my terrain and then set all the detail values to the lowest possible which makes the map less resolution and deletes the trees in the terrain.
I now need to get rid of the first terrain, meshes and cars and can achieve that by tagging them as the terrain layer. I then go into the mini map cameras properties and make sure it doesn't show objects with layer tag 'Terrain' which is done simply buy un-checking Terrain in the culling mask and vice-versa for the main camera.
Now because I made the cars apart of the terrain layer they aren't visible in the mini map now. This will not do. To represent the cars I will use plains which I will place above the cars and colour code them the same as the cars. I will add them to the mini-map layer making them only visible in the mini-map camera and then child them to their retrospective cars.
Doing all this I was able to achieve a fairly good mini map however this isn't the most professional approach to creating one but it works fine and I am happy with it. Below is a picture showing you what the map looks like.

