Multi-Robot Systems Research Group

This paper discusses Design, Implementation and Control of Series Elastic Actuator (SEA). This model was mainly dependent on usage of mechanical off-shelf components which are weigh cheaper to purchasable models developed by companies. For an appropriate performance-to-cost component selection, cost-constrained design requires careful evaluation of important performance criteria. In addition, we used These models to create force and position tracking controllers that combine PID and disturbance observer control architectures model and practical based. SEA can be used as a compliant actuator, We have been keen to utilize a collaborative robot and tackle the safety issues that arise when robots and humans interact as well as the cost issue. As a result, The goal of this study is to show how to design and install hardware as well as implementing some software programs to produce a fully functioned University of Texas Series Elastic Actuator (UT-SEA).

Intersection management is one of the most important research directions in traffic control. Autonomous vehicles are considered the near future solution to enhance the traffic systems in terms of safety, travel delay , fuel consumption reduction and passenger comfort. This paper uses Fuzzy logic control in a decentralised manner to manage a four-way intersection. The controller was tested to validate different scenarios. It succeeds to manage the intersection up to 3384 vehicle per hour. Compared to the intelligent driver model implemented by Zhang to mimic the traffic light system , the fuzzy logic controller outperforms it in terms of intersection throughput and average travel time. the intersection throughput was increased by 118.75 percent and The average travel time was reduced by 10.36 percent.

This paper proposes a three layer architecture as well as an on-board low level controller for indoor navigation of multirotors, with implementation and testing on a hexacopter. Robot Operating System (ROS) and Serial Peripheral Interface (SPI) were used to communicate between the system’s architectural levels. A decentralized cascade Proportional Integral Derivative (PID) control system was designed as a low level controller for normal operation with a fail-safe backup mode of operation to prevent a system crash in case of an unnatural architectural failure, and were implemented in the on-board computer. The system communication was highly reliable and accurate, with full communication cycle frequency of 29Hz. The simulation results are satisfactory as the system has some minor overshoot. The roll, pitch and yaw controllers are decent as they stabilize quickly with and with almost no oscillations.

For an autonomous vehicle driving on a public road, the safety of the passengers and the efficiency of the trip taken are prioritized causing the main function of the autonomous vehicle to be interpreting and inferring the intention of surrounding vehicles, and warning the driver accordingly. Recent Advanced Driving Assistance Systems (ADAS) are capable of and usually limited to, support features like forward-collision warnings, alerting the driver of hazardous road conditions, detecting road markings, and warning the driver if they are changing lanes. However, modern ADAS are still unable to perform basic vehicle-behavior-prediction humans are capable of. In this paper, we introduce and compare the results of two different methodologies, Recurrent Neural Networks (RNN) and Long Short-Term Memory networks (LSTM), for lane-changing intention prediction of surrounding vehicles. For the LSTM model, the F1-score achieved was 0.944 for lane-keeping, 0.781 for left lane-changing, and 0.942 for right lane-changing. The RNN-based model reached an F1-score of 0.704 for lane-keeping, 0.533 for left lane-changing, and 0.714 for right lane-changing. The training process of these data-driven based methodologies can be implemented using sequences of changing centroids of vehicles along with the frames and labeling of the maneuvers introduced by the PREVENTION dataset.

Multi-depot Vehicle Routing Problem is one of the most important and challenging variations of the classical Vehicle Routing Problem, where the goal is to find the routes for a fleet of vehicles to serve a number of customers, travelling from and to several depots. Due to the complexity of solving such problems, meta-heuristic algorithms are used. The Most Valuable Player algorithm is a recent technique used to solve continuous optimization problems. This study uses the Genetic Algorithm and the Ant Colony Optimization to solve the Multi-Depot Vehicle Routing Problem. A Hybrid Most Valuable Player algorithm is also proposed to solve the multi-depot vehicle routing problem. The algorithm was tested on 10 different problems and compared to two well-known techniques, Genetic Algorithm and Ant Colony Optimization. Results of the developed algorithm were satisfactory for small sized problems, however Genetic Algorithm surpassed both other algorithms in most test cases.