Search

Localization based on scalar field map matching (e.g., using gravity anomaly, magnetic anomaly, topographics, or olfaction maps) is a potential solution for navigating in Global Navigation Satellite System (GNSS)-denied environments. In this paper, a scalable framework is presented for cooperatively localizing a group of agents based on map matching given a prior map modeling the scalar field. In order to satisfy the communication constraints, each agent in the group is assigned to different subgroups. A locally centralized cooperative localization method is performed in each subgroup to estimate the poses and covariances of all agents inside the subgroup. Each agent in the group, at the same time, could belong to multiple subgroups, which means multiple pose and covariance estimates from different subgroups exist for each agent. The improved pose estimate for each agent at each time step is then solved through an information fusion algorithm. The proposed algorithm is evaluated with two different types of scalar field based simulations. The simulation results show that the proposed algorithm is able to deal with large group sizes (e.g., 128 agents), achieve 10-m level localization performance with 180 km traveling distance, while under restrictive communication constraints.

Due to the epidemic threat, more and more companies decide to automate their production lines. Given the lack of adequate security or space, in most cases, such companies cannot use classic production robots. The solution to this problem is the use of collaborative robots (cobots). However, the required equipment (force sensors) or alternative methods of detecting a threat to humans are usually quite expensive. The article presents the practical aspect of collision detection with the use of a simple neural architecture. A virtual force and torque sensor, implemented as a neural network, may be useful in a team of collaborative robots. Four different approaches are compared in this article: auto-regressive (AR), recurrent neural network (RNN), convolutional long short-term memory (CNN-LSTM) and mixed convolutional LSTM network (MC-LSTM). These architectures are analyzed at different levels of input regression (motor current, position, speed, control velocity). This sensor was tested on the original CURA6 robot prototype (Cooperative Universal Robotic Assistant 6) by Intema. The test results indicate that the MC-LSTM architecture is the most effective with the regression level set at 12 samples (at 24 Hz). The mean absolute prediction error obtained by the MC-LSTM architecture was approximately 22 Nm. The conducted external test (72 different signals with collisions) shows that the presented architecture can be used as a collision detector. The MC-LSTM collision detection f1 score with the optimal threshold was 0.85. A well-developed virtual sensor based on such a network can be used to detect various types of collisions of cobot or other mobile or stationary systems operating on the basis of human-machine interaction.