Regarding the issues of non-smooth paths, unstable obstacle avoidance, and getting stuck in local optima in mobile robot path planning, a new adaptive crayfish optimization algorithm (ACOA) based on the traditional crayfish optimization algorithm (COA) is proposed. The algorithm improves robot path planning. First, a Piecewise chaotic map is used to set up the population. It adds diversity and randomness to the population and helps the algorithm search better globally. Next, an adaptive temperature adjustment is used. This helps crayfish change their behaviors to fit the situation and balances global planning and local searching to speed up the convergence of the algorithm. Finally, a special fitness function is designed. It helps robots avoid obstacles better and creates smoother paths. Tests show that the improved algorithm works well in different complex environments. It gives shorter paths, faster convergence, and fewer turns. Compared with the COA algorithm, the ACOA algorithm reduces the path length by 3.9%, 5.3%, and 17.3% in the three maps, and the inflection points are reduced by 40%, 33%, and 45%, respectively. It also smooths paths and better avoids obstacles.