OptiState: State Estimation of Legged Robots using Gated Networks with Transformer-based Vision and Kalman Filtering

State estimation for legged robots is challenging due to their highly dynamic motion and limitations imposed by sensor accuracy. By integrating Kalman filtering, optimization, and learning-based modalities, we propose a hybrid solution that combines proprioception and exteroceptive information for estimating the state of the robot's trunk. Leveraging joint encoder and IMU measurements, our Kalman filter is enhanced through a single-rigid body model that incorporates ground reaction force control outputs from convex Model Predictive Control optimization. The estimation is further refined through Gated Recurrent Units, which also considers semantic insights and robot height from a Vision Transformer autoencoder applied on depth images. This framework not only furnishes accurate robot state estimates, including uncertainty evaluations, but can minimize the nonlinear errors that arise from sensor measurements and model simplifications through learning. The proposed methodology is evaluated in hardware using a quadruped robot on various terrains, yielding a 65% improvement on the Root Mean Squared Error compared to our VIO SLAM baseline. Code example: https://github.com/AlexS28/OptiState

SCALER: Versatile Multi-Limbed Robot for Free-Climbing in Extreme Terrains

This paper presents SCALER, a versatile free-climbing multi-limbed robot that is designed to achieve tightly coupled simultaneous locomotion and dexterous grasping. Although existing quadruped-limbed robots have shown impressive dexterous skills such as object manipulation, it is essential to balance power-intensive locomotion and dexterous grasping capabilities. We design a torso linkage and a parallel-serial limb to meet such conflicting skills that pose unique challenges in the hardware designs. SCALER employs underactuated two-fingered GOAT grippers that can mechanically adapt and offer 7 modes of grasping, enabling SCALER to traverse extreme terrains with multi-modal grasping strategies. We study the whole-body approach, where SCALER uses its body and limbs to generate additional forces for stable grasping with environments, further enhancing versatility. Furthermore, we improve the GOAT gripper actuation speed to realize more dynamic climbing in a closed-loop control fashion. With these proposed technologies, SCALER can traverse vertical, overhang, upside-down, slippery terrains, and bouldering walls with non-convex-shaped climbing holds under the Earth's gravity.

Tactile Tool Manipulation

Humans can effortlessly perform very complex, dexterous manipulation tasks by reacting to sensor observations. In contrast, robots can not perform reactive manipulation and they mostly operate in open-loop while interacting with their environment. Consequently, the current manipulation algorithms either are inefficient in performance or can only work in highly structured environments. In this paper, we present closed-loop control of a complex manipulation task where a robot uses a tool to interact with objects. Manipulation using a tool leads to complex kinematics and contact constraints that need to be satisfied for generating feasible manipulation trajectories. We first present an open-loop controller design using Non-Linear Programming (NLP) that satisfies these constraints. In order to design a closed-loop controller, we present a pose estimator of objects and tools using tactile sensors. Using our tactile estimator, we design a closed-loop controller based on Model Predictive Control (MPC). The proposed algorithm is verified using a 6 DoF manipulator on tasks using a variety of objects and tools. We verify that our closed-loop controller can successfully perform tool manipulation under several unexpected contacts. Video summarizing this work and hardware experiments are found https://youtu.be/VsClK04qDhk.

Real-to-Sim: Deep Learning with Auto-Tuning to Predict Residual Errors using Sparse Data

Achieving highly accurate kinematic or simulator models that are close to the real robot can facilitate model-based controls (e.g., model predictive control or linear-quadradic regulators), model-based trajectory planning (e.g., trajectory optimization), and decrease the amount of learning time necessary for reinforcement learning methods. Thus, the objective of this work is to learn the residual errors between a kinematic and/or simulator model and the real robot. This is achieved using auto-tuning and neural networks, where the parameters of a neural network are updated using an auto-tuning method that applies equations from an Unscented Kalman Filter (UKF) formulation. Using this method, we model these residual errors with only small amounts of data - a necessity as we improve the simulator/kinematic model by learning directly from hardware operation. We demonstrate our method on robotic hardware (e.g., manipulator arm), and show that with the learned residual errors, we can further close the reality gap between kinematic models, simulations, and the real robot.

Feasibility Study of LIMMS, A Multi-Agent Modular Robotic Delivery System with Various Locomotion and Manipulation Modes

The logistics of transporting a package from a storage facility to the consumer's front door usually employs highly specialized robots often times splitting sub-tasks up to different systems, e.g., manipulator arms to sort and wheeled vehicles to deliver. More recent endeavors attempt to have a unified approach with legged and humanoid robots. These solutions, however, occupy large amounts of space thus reducing the number of packages that can fit into a delivery vehicle. As a result, these bulky robotic systems often reduce the potential for scalability and task parallelization. In this paper, we introduce LIMMS (Latching Intelligent Modular Mobility System) to address both the manipulation and delivery portion of a typical last-mile delivery while maintaining a minimal spatial footprint. LIMMS is a symmetrically designed, 6 degree of freedom (DoF) appendage-like robot with wheels and latching mechanisms at both ends. By latching onto a surface and anchoring at one end, LIMMS can function as a traditional 6-DoF manipulator arm. On the other hand, multiple LIMMS can latch onto a single box and behave like a legged robotic system where the package is the body. During transit, LIMMS folds up compactly and takes up much less space compared to traditional robotic systems. A large group of LIMMS units can fit inside of a single delivery vehicle, opening the potential for new delivery optimization and hybrid planning methods never done before. In this paper, the feasibility of LIMMS is studied and presented using a hardware prototype as well as simulation results for a range of sub-tasks in a typical last-mile delivery.

Multi-Modal Multi-Agent Optimization for LIMMS, A Modular Robotics Approach to Delivery Automation

In this paper we present a motion planner for LIMMS, a modular multi-agent, multi-modal package delivery platform. A single LIMMS unit is a robot that can operate as an arm or leg depending on how and what it is attached to, e.g., a manipulator when it is anchored to walls within a delivery vehicle or a quadruped robot when 4 are attached to a box. Coordinating amongst multiple LIMMS, when each one can take on vastly different roles, can quickly become complex. For such a planning problem we first compose the necessary logic and constraints. The formulation is then solved for skill exploration and can be implemented on hardware after refinement. To solve this optimization problem we use alternating direction method of multipliers (ADMM). The proposed planner is experimented under various scenarios which shows the capability of LIMMS to enter into different modes or combinations of them to achieve their goal of moving shipping boxes.

Learning Near-global-optimal Strategies for Hybrid Non-convex Model Predictive Control of Single Rigid Body Locomotion

Convex model predictive controls (MPCs) with a single rigid body model have demonstrated strong performance on real legged robots. However, convex MPCs are limited by their assumptions such as small rotation angle and pre-defined gait, limiting the richness of potential solutions. We remove those assumptions and solve the complete mixed-integer non-convex programming with single rigid body model. We first collect datasets of pre-solved problems offline, then learn the problem-solution map to solve this optimization fast for MPC. If warm-starts can be found, offline problems can be solved close to the global optimality. The proposed controller is tested by generating various gaits and behaviors depending on the initial conditions. Hardware test demonstrates online gait generation and adaptation running at more than 50 Hz based on sensor feedback.

Auto-Calibrating Admittance Controller for Robust Motion of Robotic Systems

We demonstrate an admittance controller with auto-tuning that can be applied for single and multi-point contact robots (e.g., legged robots with point feet or multi-finger grippers). The controller's objective is to track wrench profiles of each contact point while considering the additional torque due to rotational friction. Our admittance controller is adaptive during online operation by using an auto-tuning method that tunes the gains of the controller while following several training objectives that facilitate controller stability, such as tracking the wrench profile as closely as possible, ensuring control outputs that are within force limits that minimize slippage, and avoids kinematic singularity. We demonstrate the robustness of our controller on hardware for both manipulation and locomotion tasks using a multi-limbed climbing robot.

Simultaneous Contact-Rich Grasping and Locomotion via Distributed Optimization Enabling Free-Climbing for Multi-Limbed Robots

While motion planning of locomotion for legged robots has shown great success, motion planning for legged robots with dexterous multi-finger grasping is not mature yet. We present an efficient motion planning framework for simultaneously solving locomotion (e.g., centroidal dynamics), grasping (e.g., patch contact), and contact (e.g., gait) problems. To accelerate the planning process, we propose distributed optimization frameworks based on Alternating Direction Methods of Multipliers (ADMM) to solve the original large-scale Mixed-Integer NonLinear Programming (MINLP). The resulting frameworks use Mixed-Integer Quadratic Programming (MIQP) to solve contact and NonLinear Programming (NLP) to solve nonlinear dynamics, which are more computationally tractable and less sensitive to parameters. Also, we explicitly enforce patch contact constraints from limit surfaces with micro-spine grippers. We demonstrate our proposed framework in the hardware experiments, showing that the multi-limbed robot is able to realize various motions including free-climbing at a slope angle 45{\deg} with a much shorter planning time.

SCALER: A Tough Versatile Quadruped Free-Climber Robot

This paper introduces SCALER, a quadrupedal robot that demonstrates climbing on bouldering walls, overhangs, and ceilings and trotting on the ground. SCALER is one of the first high-degrees of freedom four-limbed robots that can free-climb under the Earth's gravity and one of the most mechanically efficient quadrupeds on the ground. Where other state-of-the-art climbers are specialized in climbing itself, SCALER promises practical free-climbing with payload \textit{and} ground locomotion, which realizes true versatile mobility. A new climbing gait, SKATE gait, increases the payload by utilizing the SCALER body linkage mechanism. SCALER achieves a maximum normalized locomotion speed of $1.87$ /s, or $0.56$ m/s on the ground and $1.2$ /min, or $0.42$ m/min in bouldering wall climbing. Payload capacity reaches $233$ % of the SCALER weight on the ground and $35$ % on the vertical wall. Our GOAT gripper, a mechanically adaptable underactuated two-finger gripper, successfully grasps convex and non-convex objects and supports SCALER.