These small simplified images representing things are called pictograms (they're sometimes called icons but that also makes implications about how they are used). See also What do you call these infographic icons? which discusses a different style of the same thing.
You can browse thousands and thousands of pictograms like that at the noun project, and also upload and give away/sell your own.
For the style, there aren't really fixed names for different styles: pictograms are so stripped down there are only a few variables anyway, so it's easier to just describe it:
- How many colours: these are single-colour or monochrome. Pictograms are most often monochrome (all the ones on the noun project are) but you also quite often see them with restricted palettes of 2-5 colours.
- Line art / outlined, or solid shape / silhouette, or (rare) a combination of the two.
- Level of detail: these are medium-high detail, there's texture but only a little bit.
- Tone or mood: e.g. cartoony, instructional, soft, hard, smooth, crisp, frivolous, sober...
So I'd describe these as monochrome line art pictograms with a medium-high level of detail and a slightly cartoony, smooth but sober style.
To add to what others have said, use a vector drawing application like Illustrator to create these, unless you're doing pixel art. A few more Illustrator tools to use:
- Pen tool, obviously. Pencil tool is good for line-art freehand drawing but for pictograms you'll want to be more tightly controlled than that
- Align window for making sure lines align perfectly, centre perfectly, and for evenly spacing things such as lines/dots that imply texture
- Pathfinder window for chopping sections out of shapes
Don't be fooled by the easy, simple technology, though. The tech is simple but they're hard to do well: unless you're making really recognisable shapes, it takes a lot of skill to pare something down to the bare essentials and make it instantly recognisable even at tiny sizes.
It's easy to get carried away adding fragile details that actually just get lost and only add complexity without aiding recognition. It helps to use fat chunky grids, and to zoom out/step back often.