No abstract

Abstract-Affordances encode relationships between actions, objects, and effects. They play an important role on basic cognitive capabilities such as prediction and planning. We address the problem of learning affordances through the interaction of a robot with the environment, a key step to understand the world properties and develop social skills. We present a general model for learning object affordances using Bayesian networks integrated within a general developmental architecture for social robots. Since learning is based on a probabilistic model, the approach is able to deal with uncertainty, redundancy, and irrelevant information. We demonstrate successful learning in the real world by having an humanoid robot interacting with objects. We illustrate the benefits of the acquired knowledge in imitation games.

Brain-machine interfaces (BMI) usually decode movement parameters from cortical activity to control neuroprostheses. This requires subjects to learn to modulate their brain activity to convey all necessary information, thus imposing natural limits on the complexity of tasks that can be performed. Here we demonstrate an alternative and complementary BMI paradigm that overcomes that limitation by decoding cognitive brain signals associated with monitoring processes relevant for achieving goals. In our approach the neuroprosthesis executes actions that the subject evaluates as erroneous or correct, and exploits the brain correlates of this assessment to learn suitable motor behaviours. Results show that, after a short user’s training period, this teaching BMI paradigm operated three different neuroprostheses and generalized across several targets. Our results further support that these error-related signals reflect a task-independent monitoring mechanism in the brain, making this teaching paradigm scalable. We anticipate this BMI approach to become a key component of any neuroprosthesis that mimics natural motor control as it enables continuous adaptation in the absence of explicit information about goals. Furthermore, our paradigm can seamlessly incorporate other cognitive signals and conventional neuroprosthetic approaches, invasive or non-invasive, to enlarge the range and complexity of tasks that can be accomplished.

Abstract. Inverse reinforcement learning addresses the general problem of recovering a reward function from samples of a policy provided by an expert/demonstrator. In this paper, we introduce active learning for inverse reinforcement learning. We propose an algorithm that allows the agent to query the demonstrator for samples at specific states, instead of relying only on samples provided at "arbitrary" states. The purpose of our algorithm is to estimate the reward function with similar accuracy as other methods from the literature while reducing the amount of policy samples required from the expert. We also discuss the use of our algorithm in higher dimensional problems, using both Monte Carlo and gradient methods. We present illustrative results of our algorithm in several simulated examples of different complexities.

The closed-loop control of rehabilitative technologies by neural commands has shown a great potential to improve motor recovery in patients suffering from paralysis. Brain–machine interfaces (BMI) can be used as a natural control method for such technologies. BMI provides a continuous association between the brain activity and peripheral stimulation, with the potential to induce plastic changes in the nervous system. Paraplegic patients, and especially the ones with incomplete injuries, constitute a potential target population to be rehabilitated with brain-controlled robotic systems, as they may improve their gait function after the reinforcement of their spared intact neural pathways. This paper proposes a closed-loop BMI system to control an ambulatory exoskeleton—without any weight or balance support—for gait rehabilitation of incomplete spinal cord injury (SCI) patients. The integrated system was validated with three healthy subjects, and its viability in a clinical scenario was tested with four SCI patients. Using a cue-guided paradigm, the electroencephalographic signals of the subjects were used to decode their gait intention and to trigger the movements of the exoskeleton. We designed a protocol with a special emphasis on safety, as patients with poor balance were required to stand and walk. We continuously monitored their fatigue and exertion level, and conducted usability and user-satisfaction tests after the experiments. The results show that, for the three healthy subjects, 84.44 ± 14.56% of the trials were correctly decoded. Three out of four patients performed at least one successful BMI session, with an average performance of 77.6 1 ± 14.72%. The shared control strategy implemented (i.e., the exoskeleton could only move during specific periods of time) was effective in preventing unexpected movements during periods in which patients were asked to relax. On average, 55.22 ± 16.69% and 40.45 ± 16.98% of the trials (for healthy subjects and patients, respectively) would have suffered from unexpected activations (i.e., false positives) without the proposed control strategy. All the patients showed low exertion and fatigue levels during the performance of the experiments. This paper constitutes a proof-of-concept study to validate the feasibility of a BMI to control an ambulatory exoskeleton by patients with incomplete paraplegia (i.e., patients with good prognosis for gait rehabilitation).