• Task: 

  • Layout optimization of a battery disassembling robotic cell in order to minimize the cycle time. 

  • The layout will be used for the new I2FORD cell at PISA lab in Brescia.

• Software: ROS2, C++

• References: 

  • Villagrossi, E., Dinon, T. Robotics for electric vehicles battery packs disassembly towards sustainable remanufacturing. Jnl Remanufactur

 AVAILABLE

• Background: The I2FORD cell includes a 2-axis cart used to move the battery in the cell and to allow the robots to reach every side of battery.

• Task: 

  • Develop an hardware interface for the battery handler cart compatible with ROS 2. 

  • The interface will communicate with a low level drive, in control of two axis (one to control the height of the battery, the other to control the inclination)

  • Communication will happen through EtherCAT bus.

• Software: ROS2, ros2_control, EtherCAT ROS2 Driver, C++

• References:

  • Villagrossi, E., Dinon, T. Robotics for electric vehicles battery packs disassembly towards sustainable remanufacturing. Jnl Remanufactur

UNAVAILABLE

• Background: The smart radars by Inxpect detect moving objects in their field of view and estimate their position and angle with respect to the sensor itself. This is particularly useful for detecting humans and stopping or slowing down robots when specific safety areas are crossed. 
  Multiple radars are used to create virtual fences around the robots, eliminating physical barriers and allowing for a faster restart of operations when the critical area becomes free again. The challenge consists of calibrating the array of sensors with respect to a common reference frame given noisy sensor measurements. 

• Task: 

  • Expand the ROS2-python package used to read the sensor status via a MODBUS interface (Ethernet interface for real-time data monitoring). Explore other interfaces such as FSoE (Ethernet-based safety Fieldbus - Safety over EtherCAT®). 

  • Simultaneous calibration of multiple Inxpect radars. 

  • Integration of the radar information in the safety routines of the robotic cell.

• Software: ROS2, Python

• References:

  • https://www.inxpect.com/en/products

AVAILABLE

• Background: Speed and Separation Monitoring (SSM) is proposed in ISO/TS-15066 as a safety method for human-robot collaborative cells. It requires a minimum separation distance between the human and the robot to avoid collisions, which translates into a maximum robot velocity based on the current position and velocity of both.

• Task: 

  • Estimation of future values of SSM velocity scaling using Artificial Intelligence techniques (e.g. Neural Network).

  • Assess the possibility of integrating such methodology in a replanning routine for collaborative robots.

• Software: MATLAB (Python)

• References:

  • C. Tonola, M. Faroni, M. Beschi and N. Pedrocchi, "Anytime Informed Multi-Path Replanning Strategy for Complex Environments," in IEEE Access

  • Marvel, J. A., & Norcross, R. (2017). Implementing speed and separation monitoring in collaborative robot workcells. Robotics and computer-integrated manufacturing

AVAILABLE

• Background: Speed and Separation Monitoring (SSM) is proposed in ISO/TS-15066 as a safety method for human-robot collaborative cells. It requires a minimum separation distance between the human and the robot to avoid collisions, which translates into a maximum robot velocity based on the current position and velocity of both.
  Elastic Strip is an obstacle avoidance method that adds an elastic term to the control action to allow temporary deviations from the original trajectory.

• Task: 

  • Include SSM computations into the Elastic Strip algorithm. The idea is to exploit the elastic strip to bend the trajectory to maximize the SSM and thus avoid slowdowns.

• Software: MATLAB (C++, Python)

• References:

  • http://robotics.stanford.edu/~oli/elastic.html

  • Marvel, J. A., & Norcross, R. (2017). Implementing speed and separation monitoring in collaborative robot workcells. Robotics and computer-integrated manufacturing

AVAILABLE

• Background: The replanning procedure in a robot consists of computing a new trajectory in real time due to changes in the environment (new or moving obstacles, etc.)

• Task: 

  • Including pose constraints in the MARS replanner to plan the new path while keeping a predefined orientation of the tool.

• Software: MATLAB (C++)

• References:

  • C. Tonola, M. Faroni, M. Beschi and N. Pedrocchi, "Anytime Informed Multi-Path Replanning Strategy for Complex Environments," in IEEE Access

  • C. Tonola, M. Beschi, M. Faroni and N. Pedrocchi, "OpenMORE: an open-source tool for sampling-based path replanning in ROS," 2023 IEEE 28th International Conference on Emerging Technologies and Factory Automation (ETFA)

AVAILABLE

• Background: Real-time tracking of the operator enables safer and more efficient cooperation between the human and the robot while performing a collaborative task.
  The robot speed is scaled based on the distance to the operator's body according to the SSM strategy. However, to improve fluency and reduce idle time, we seek to predict the future movement of the human, so that the robot can replan in advance its path.

• Task: 

  • Test the following Python libraries for time-series analysis: sktime, tsfresh.

  • Explore the Inverse Optimal Control [1] strategy: the human is modeled as an optimal controller that maximizes a merit function learned from demonstrations. The goal is to model and learn the parameters of this function.

• Software: Python (MATLAB)

• References:

  • [1] Example application

    • Paper: Kitani, K. M., Ziebart, B. D., Bagnell, J. A., & Hebert, M. (2012). Activity forecasting. In Computer Vision–ECCV 2012: 12th European Conference on Computer Vision, Florence, Italy, October 7-13, 2012, Proceedings, Part IV 12 (pp. 201-214). Springer Berlin Heidelberg. 

    • Github (unofficial): https://github.com/thirakawa/ActivityForecastingPython.git

AVAILABLE

• Background: In the context of robot motion replanning, one can adopt a reactive or a proactive approach. The former consists of first predicting the operator's behavior and subsequently optimizing the path of the robot based on this prediction. The latter tries to jointly optimize the motion of both agents, the human and the robot, using Differential Game Theory. This is done along a receding horizon to update the likely behavior of the operator based on new data from the skeleton-tracking module.

• Task: 

  • Investigate the Iterative Linear Quadratic Regulator (ILQR) [1, 2].

  • Extend the ilqgames library [3] to the case of human-robot interaction in a collaborative cell. Devise tests to compare this replanning  strategy with existing approaches.

• Software: C++, ROS/ROS2

• References:

  • [1] Li, W., & Todorov, E. (2004, August). Iterative linear quadratic regulator design for nonlinear biological movement systems. In First International Conference on Informatics in Control, Automation and Robotics (Vol. 2, pp. 222-229). SciTePress.

  • [2] Tassa, Y., Mansard, N., & Todorov, E. (2014, May). Control-limited differential dynamic programming. In 2014 IEEE International Conference on Robotics and Automation (ICRA) (pp. 1168-1175). IEEE. 

  • [3] Fridovich-Keil, D., Ratner, E., Peters, L., Dragan, A. D., & Tomlin, C. J. (2020, May). Efficient iterative linear-quadratic approximations for nonlinear multi-player general-sum differential games. In 2020 IEEE international conference on robotics and automation (ICRA) (pp. 1475-1481). IEEE.

AVAILABLE