This project introduces PyTeLo, a modular and versatile Python-based software that facilitates working with temporal logic languages, specifically MTL, STL, and wSTL. Applying PyTeLo requires only a string representation of the temporal logic specification and, optionally, the system’s dynamics of interest. Next, PyTeLo reads the specification using an ANTLR-generated parser and generates an Abstract Syntax Tree (AST) that captures the structure of the formula. For synthesis, the AST recursively encodes the specification into a Mixed Integer Linear Program (MILP) that is solved using a commercial solver such as Gurobi.

We’re building a physical human–humanoid interaction (pHHI) stack that lets a general-purpose humanoid safely share load, guidance, and balance with a person. Our approach combines online intent estimation (derived from wrist/arm force–torque and vision) with whole-body model-predictive control and variable-impedance behaviors, while enforcing passivity, friction cones, and ergonomic force/torque limits. A “compliance scheduler” arbitrates shared autonomy, being soft when guided, firm when supporting, and proactive to prevent a loss of balance. We’ll validate on-hand walking, sit-to-stand assist, and dual-arm co-manipulation. The primary goal of this project is to develop a reproducible pHHI toolkit that transforms a humanoid into a reliable physical teammate. 

This project enhances high-level decision planning and control synthesis of autonomous systems by introducing a novel approach to Partial Satisfaction (PS) of Temporal Logic specifications.

Our framework, which accepts mission specifications in STL, MTL, and wSTL, addresses a significant challenge in the field. Since specifications are typically humanly designed, the potential for defining conflicting and competing subformulae and even an overall unfeasible specification is very high.

The primary goal of this project is to maximize the fulfillment of complex missions by optimizing the satisfaction of STL subformulas, emphasizing prioritizing those with greater significance or preferences.

The approach involves a bi-level optimization framework, where the inner level captures the partial satisfiability of the specification using mixed-integer linear programming (MILP) methods. The outer optimization level computes the robustness of satisfiable predicates, and it is then addressed using a linear program, leveraging the inner solutions for approximation.

This project focuses on optimizing the planning and coordination of multi-robot systems by leveraging the power of mixed-integer linear programming (MILP) combined with temporal logic. We exploit the efficiency of network flow optimization problems to provide advanced planning methodologies that enable multi-robot systems to perform complex, coordinated tasks with greater precision and adaptability.

1. Modular Aerial Robots: Modular aerial robots equipped with specialized tools to accomplish specific tasks. These robots can reconfigure themselves to achieve optimal configurations for subsequent tasks or adapt their movements based on environmental demands, enhancing their overall mission performance.

2. Heterogeneous Robotic Teams: Coordinating heterogeneous robots with varying capabilities, focusing on efficient resource transport and management. The research considers multiple resource types—both divisible and indivisible—and explores different transportation methods, including uniform and compartmental capacities, tailored to the specific resource types.

3. Swarm Dynamics: The behavior of robotic swarms that can split and merge within an environment. These swarms are designed to monitor different regions while ensuring sufficient agent presence during the mission, guaranteeing continuous coverage and adaptability to changing conditions.

Papers on planning for multi-robot systems:

[1] Planning for modular aerial robotic tools with temporal logic constraints (CDC-2022)

[2] Temporal logic swarm control with splitting and merging (ICRA-2023)

[3] Planning for heterogeneous teams of robots with temporal logic, capability, and resource constraints (IJRR-2024) 

[4] An Iterative Approach for Heterogeneous Multi-Agent Route Planning with Temporal Logic Goals and Travel Duration Uncertainty (ICRA-2024) 

[5] An Iterative Approach for Heterogeneous Multi-Agent Route Planning with Resource Transportation Uncertainty and Temporal Logic Goals 

[6] Truck Fleet Coordination for Warehouse Trailer Management by Temporal Logic with Energy Constraints 

This project focuses on the cooperative manipulation and transportation of objects using multiple quadrotors connected by cables. The goal is to develop a system where multiple quadrotors can work together to lift, maneuver, and transport objects that would be too large or heavy for a single drone to handle. 

There have been multiple approaches considered:

1. Multiple quadrotors are attached to the same point of a load with cables that can be stretchable or not. Quadrotors transport the load while minimizing the control effort.

2. Multiple Catenary robots (quadrotors attached by a cable) drag, roll, or lift a box depending on the level of friction experienced on the floor.

Papers on cooperative aerial manipulation:

[1] Cooperative transportation of a cable-suspended load by multiple quadrotors (NECSYS-2019)

[2] Robust Adaptive Synchronization of Interconnected Heterogeneous Quadrotors Transporting a Cable-Suspended Load (ICRA-2021)

[3] Adaptive Multi-Quadrotor Control for Cooperative Transportation of a Cable-Suspended Load (ECC-2021)

[4] The catenary robot: Design and control of a cable propelled by two quadrotors (IEEE-RAL-2021)

[5] Non-prehensile manipulation of cuboid objects using a catenary robot (IROS-2021)

[6] Adaptive control for cooperative aerial transportation using catenary robots (AIRPHARO-2021)