When people learn to make decisions from experience, a reasonable intuition is that additional relevant information should improve their performance. In contrast, we find that additional information about foregone rewards (i.e., what could have gained at each point by making a different choice) severely hinders participants’ ability to repeatedly make choices that maximize long-term gains. We conclude that foregone reward information accentuates the local superiority of short-term options (e.g., consumption) and consequently biases choice away from productive long-term options (e.g., exercise). These conclusions are consistent with a standard reinforcement-learning mechanism that processes information about experienced and forgone rewards. In contrast to related contributions using delay-of-gratification paradigms, we do not posit separate top-down and emotion-driven systems to explain performance. We find that individual and group data are well characterized by a single reinforcement-learning mechanism that combines information about experienced and foregone rewards.

Visual simultaneous localization and mapping (visual SLAM) has been well developed in recent decades. To facilitate tasks such as path planning and exploration, traditional visual SLAM systems usually provide mobile robots with the geometric map, which overlooks the semantic information. To address this problem, inspired by the recent success of the deep neural network, we combine it with the visual SLAM system to conduct semantic mapping. Both the geometric and semantic information will be projected into the 3D space for generating a 3D semantic map. We also use an optical-flow-based method to deal with the moving objects such that our method is capable of working robustly in dynamic environments. We have performed our experiments in the public TUM dataset and our recorded office dataset. Experimental results demonstrate the feasibility and impressive performance of the proposed method.

This paper presents a new approach to the management of the environmental map for mobile robots in dynamic environments. The environmental map is built of primitive features, such as lines, points, and even circles, extracted from ambiguous data captured by the robot's sonar sensor ring. The feature map must be managed because the indoor surroundings where mobile robots operate are continuously changing due to nonstationary objects, such as wastebaskets, tables, and people. The features are processed by trimming, division, or removal, depending on the dynamic circumstances. All processing refers to the occupancy probabilities of grid squares generated for the map features. The occupancy probabilities of the squares are updated using the Bayesian updating model with the sonar sensor data. Experimental results demonstrate the validity of the proposed method.

Motion planning for manipulators with many degrees of freedom is a complex task. The research in this area has been mostly restricted to static environments. This paper presents a comparative analysis of three reactive on-line path-planning methods for manipulators: the elastic-strip, strategy-based and potential field methods. Both the elastic-strip method [O. Brock and O. Khatib, “Elastic strips: A framework for integrated planning and execution,” Int. Symp. Exp. Robot. 245–254 (1999)] and the potential field method [O. Khatib, “Real-time obstacle avoidance for manipulators and mobile robots,” Int. J. Robot. Res.5(1), 90–98 (1986)] have been adapted by the authors to the problem at hand related to our multi-manipulator system (MMS) (three manipulators with five degrees of freedom each). Strategy-based method is an original contribution by the authors [M. Mediavilla, J. L. González, J. C. Fraile and J. R. Perán, “Reactive approach to on-line path planning for robot manipulators in dynamic environments,” Robotica20, 375–384 (2002); M. Mediavilla, J. C. Fraile, T. González and I. J. Galindo, “Selection of strategies for collision-free motion in multi-manipulator systems,” J. Intell. Robot Syst38, 85–104 (2003)].

The three methods facilitate on-line path planning for our MMS in dynamic environments with collision avoidance, where the three manipulators may move at the same time in their common workspace. We have defined some ‘basic motion problems’ for the MMS, and a series of simulations has been running that will tell us how effective each path-planning method is. The simulations have been performed and the obtained results have been analysed by using a software program developed by the authors.

The paper also presents experimental results obtained applying the path-planning methods to our MMS, that perform pick-and-place tasks sharing common working areas.

This paper describes a new approach to path planning of robot manipulators with many degrees of freedom. It is designed for on-line motion in dynamic and unpredictable environments. The robots react to moving obstacles using a local and reactive algorithm restricted to a subset of its configuration space. The lack of a long-term view of local algorithms (local minima problems) is solved using an off-line pre-planning stage that chooses the subset of the configuration space that minimises the probability of not finding collision free paths. The approach is implemented and tested on a system of three Scorbot-er IX five link robots.