Object detection is an essential component of autonomous mobile robotic systems, enabling robots to understand and interact with the environment. Object detection and recognition have made significant progress using convolutional neural networks (CNNs). Widely used in autonomous mobile robot applications, CNNs can quickly identify complicated image patterns, such as objects in a logistic environment. Integration of environment perception algorithms and motion control algorithms is a topic subjected to significant research. On the one hand, this paper presents an object detector to better understand the robot environment and the newly acquired dataset. The model was optimized to run on the mobile platform already on the robot. On the other hand, the paper introduces a model-based predictive controller to guide an omnidirectional robot to a particular position in a logistic environment based on an object map obtained from a custom-trained CNN detector and LIDAR data. Object detection contributes to a safe, optimal, and efficient path for the omnidirectional mobile robot. In a practical scenario, we deploy a custom-trained and optimized CNN model to detect specific objects in the warehouse environment. Then we evaluate, through simulation, a predictive control approach based on the detected objects using CNNs. Results are obtained in object detection using a custom-trained CNN with an in-house acquired data set on a mobile platform and in the optimal control for the omnidirectional mobile robot.

This paper presents a passivity based torque control structure for a high power Permanent Magnet Synchronous Motor (PMSM) used in Electric Vehicles (EV). A comparative study is done with a parallel PI algorithm in terms of response time, settling time, stationary error and required input electrical energy to reach the desired operating point. Moreover, parameters variations are simulated in order to evaluate the performance of the Passivity Based Control (PBC) scheme. In both control structures, the angular speed loop is controlled through a parallel PI algorithm. First, the mathematical model of PMSM is obtained in terms of Euler-Lagrange (EL) formalism which is the starting point for the passivity based torque controller design. This is followed by the damping injection and the re-modeling of the closed loop energy storage function in order to achieve passivity and asymptotic stability. Finally, the desired currents are derived from the torque request provided in the outer speed control loop. The algorithms are numerically simulated in MATLAB-Simulink.

Permanent magnet synchronous motors (PMSM) are widely used in electric vehicle application due to their high power density, high efficiency and very good performance at low speeds. Safety concerns in automotive industry require monitoring of the system to ensure a correct and safe operation of the electric vehicle. This paper presents three software methods for estimating the winding electrical resistance and detecting an open phase failure during torque control operation of a PMSM. All three methods can be applied online. Differences between these methods are reflected into computational effort and efficiency. The described methods were numerically simulated and tested in MATLAB. The first two methods monitor the electrical resistance of the stator windings which are estimated through an Extended Kalman Filter (EKF) and an Unscented Kalman Filter (UKF). Hence, the three phase model and the rotating reference frame model are both used as internal models for the proposed Kalman filters and the estimation performance is evaluated in each case. In order to detect an open phase fault, the gradient of the estimated electrical resistance is monitored. The third method which is independent of the motor parameters calculates the Discrete Fourier Transformation (DFT) coefficients of the fundamental frequency in the phase currents signals. Hence, due to the low computational effort and good performance, the Goertzel algorithm is proposed. Conclusions are drawn based on estimation accuracy, time response and complexity of each method.

This paper presents an on-line method for determining the hottest spot temperature of the copper windings of a Permanent Magnet Synchronous Machine (PMSM). The purpose is to protect the Electrical Machine (EM) from high temperatures which reduces the lifetime of the machine and can eventually lead to insulation faults of the motor. The method relies on experimental data acquired from a network of thermocouples placed around the stator windings of an actual PMSM. A mathematical model is proposed and the parameters are derived from the measurement data. Finally, the temperature of the hottest spot is determined on-line based on a single, fixed, temperature sensor placed in an accessible area of the stator (It is assumed that this is not the hottest temperature spot) and on the operating point of the PMSM. The data is combined in a Particle Filter (PF). The described method can be used with minor changes to any type of EM. The algorithm is executed in an offline environment with experimental data from a real motor-load test bench.

In this paper, an experiment is described which involved the usage of a set-up of 4 development boards based on Ultra Wide Band transceiver DW1000: three of them are configured as anchors while the forth one is configured as a tag. It is assumed that the tag is placed on a fixed reference (e.g. a traffic sign) and the anchors are placed on a moving vehicle (e.g on three corners of the vehicle). The paper deals with the improvement of the measurement of the distance between the tag and the anchors starting from the firmware of the transceiver. The actual firmware is modified such that the measurement sequence is performed faster in order to improve the accuracy of the localization.