Behavioral cloning is an imitation learning technique that teaches an agent how to behave through expert demonstrations. Recent approaches use self-supervision of fully-observable unlabeled snapshots of the states to decode state-pairs into actions. However, the iterative learning scheme from these techniques are prone to getting stuck into bad local minima. We address these limitations incorporating a two-phase model into the original framework, which learns from unlabeled observations via exploration, substantially improving traditional behavioral cloning by exploiting (i) a sampling mechanism to prevent bad local minima, (ii) a sampling mechanism to improve exploration, and (iii) self-attention modules to capture global features. The resulting technique outperforms the previous state-of-the-art in four different environments by a large margin.

Imitation from observation is a computational technique that teaches an agent on how to mimic the behavior of an expert by observing only the sequence of states from the expert demonstrations. Recent approaches learn the inverse dynamics of the environment and an imitation policy by interleaving epochs of both models while changing the demonstration data. However, such approaches often get stuck into sub-optimal solutions that are distant from the expert, limiting their imitation effectiveness. We address this problem with a novel approach that overcomes the problem of reaching bad local minima by exploring: (i) a self-attention mechanism that better captures global features of the states; and (ii) a sampling strategy that regulates the observations that are used for learning. We show empirically that our approach outperforms the state-of-the-art approaches in four different environments by a large margin.

Battery life is yet one of the main limiting factors to a robot’s total mission time, and efficient energy management is paramount in a robotic application. In this paper, we integrate energy awareness in the path planning of a mobile robot performing autonomous navigation. Our contributions are: 1) The formalization of a planning domain for mobile robot path planning which accounts for energy consumption and integrates energy actions in the generated plans; 2) A proof of concept of automatic path planning that avoids high energy areas in a known environment. We test our approach in simulation, extending an embedded computer’s total battery discharge time by approximately 42.8%, and in a real ground mobile robot, achieving a mean energy draw reduction of 52.02%, both compared to conventional path planning.

Computer-based assistants have recently attracted much interest due to its applicability to ambient assisted living. Such assistants have to detect and recognize the high-level activities and goals performed by the assisted human beings. In this work, we demonstrate activity recognition in an indoor environment in order to identify the goal towards which the subject of the video is pursuing. Our hybrid approach (HAPRec) combines an action recognition module and a goal recognition algorithm to identify the ultimate goal of the subject in the video. The demonstration can be seen at:

Recently action recognition has been used for a variety of applications such as surveillance, smart homes, and in-home elder monitoring. Such applications usually focus on recognizing human actions without taking into account the different scenarios where the action occurs. In this paper, we propose a two-stream architecture that considers not only the movements to identify the action, but also the context scene where the action is performed. Experiments show that the scene context may improve the recognition of certain actions. Our proposed architecture is tested against baselines and the standard two-stream network.