It is essential for agents to work together with others to accomplish common missions without previous knowledge of the team-mates, a challenge known as ad-hoc teamwork. In these systems, an agent estimates the algorithm and parameters of others in an on-line manner, in order to decide its own actions for effective teamwork. Meanwhile, agents often must coordinate in a decentralised fashion to complete tasks that are displaced in an environment (e.g., in foraging, demining, rescue or fire control), where each member autonomously chooses which task to perform. By harnessing this knowledge, better estimation techniques would lead to better performance. Hence, we present On-line Estimators for Ad-hoc Task Allocation, a novel algorithm for team-mates' type and parameter estimation in decentralised task allocation. We ran experiments in the level-based foraging domain, where we obtain lower error in parameter and type estimation than previous approaches, and a significantly better performance in finishing all tasks.