To address the challenges of slow convergence, susceptibility to local optima and path redundancy in the path planning of concrete pouring robots in complex construction environments, an improved ant colony algorithm?based path planning optimization method for concrete robots is proposed. First, a new pheromone update mechanism is formulated and the hindsight experience replay (HER) algorithm is applied to define pseudo?target points, thereby addressing the slow convergence and local optimum entrapment issues of conventional ant colony algorithm (ACO). Second, a new obstacle heuristic factor is designed to improve the obstacle avoidance ability of the traditional ant colony algorithm.Third, to solve the limitation of path redundancy in the traditional ant colony algorithm, a curve smoothing function is introduced to eliminate redundant nodes and improve the path quality. Simulation experiments show that the algorithm proposed in this paper has good effectiveness and stability in terms of the shortest path length, the number of turning points and iteration efficiency.