Walking robot simulation

Link: https://leetcode.com/problems/walking-robot-simulation/

Solution:

Topics: simulation

Intuition
This is not a very difficult problem, but it has some edge cases that must be considered. My first solution was a very bloated implementation, so If anything it’s a clean implementation challenge.

Implementation

def walking_robot(commands, obstacles):
	obstacles = set([tuple(obs) for obs in obstacles])
	dirs = ((0, 1), (1, 0), (0, -1), (-1, 0))
	max_dist = 0
	curr_dir = 0
	x, y = 0, 0
	for c in commands:
		if c == -1:
			curr_dir += 1
			curr_dir %= 4
		elif c == -2:
			curr_dir -= 1
			curr_dir %= 4
		else:
			a, b = dirs[curr_dir]
			for _ in range(c):
				if (x+a, y+b) in obstacles:
					break
				x += a
				y += b
			max_dist = max(max_dist, x*x + y*y)
	return max_dist
 
#time: o(9n)
#memory: o(len(obstacles))

Review 1
Fun problem! The key to remember for clean implementation in these euclidian simulation problems is to create a directions array that whose tuple represents one unit of movement in that direction if added to the current position.

For example:

Lets say we are starting at positon (0,0), how do we go up by one unit?

add (0, 1) to the position! 

Therefore...

dirs = ((0, 1), (1, 0), (0, -1), (-1, 0))
         up      right   left     down

So if we need to move n steps down, add (-1, 0) to the current position n times!

Of course we don’t have to do it this way, but it gets bloated quick.

review