Aditya Bhatt

Aditya Bhatt, Himavarshini Yarragangu, Urvish Shah et al.

Collaborative transport of objects via pushing by multiple robots has many applications, ranging from construction and warehouse environments to post disaster debris clean-up. Achieving collaborative transport over surfaces with different inclination and friction properties however poses unique challenges. To address these challenges, this paper presents an asynchronous decentralized task and motion planning approach for transporting rectangular boxes of varying mass over flat, uphill and downhill terrain. Such a decentralized approach alleviates communication, synchronization and consensus needs and mitigates single point of failure issues. Our approach, called R2P2 or Roles with Rules and Proportional-control Primitive, assigns roles (e.g., push, support and prevent) to robots based on rules cognizant of the mode of manipulation needed (box rotation vs translation); this is followed by either rule-based control or proportional control of robot velocity based on the roles. Each robot is assumed to observe the location and heading of self and the box in executing the role and controls. R2P2 is evaluated with a six-robot team deployed in a simulator built using NVIDIA IsaacSim -- demonstrating generalizability across different surface friction/inclination and box mass scenarios, and better success rate compared to a standard virtual-leader-follower method. R2P2 is also successfully validated with a physical experiment, where it is executed onboard four turtlebots tasked with moving a 1.2 kg box.

Moritz Meser, Aditya Bhatt, Boris Belousov et al.

We tackle the recently introduced benchmark for whole-body humanoid control HumanoidBench using MuJoCo MPC. We find that sparse reward functions of HumanoidBench yield undesirable and unrealistic behaviors when optimized; therefore, we propose a set of regularization terms that stabilize the robot behavior across tasks. Current evaluations on a subset of tasks demonstrate that our proposed reward function allows achieving the highest HumanoidBench scores while maintaining realistic posture and smooth control signals. Our code is publicly available and will become a part of MuJoCo MPC, enabling rapid prototyping of robot behaviors.

Michael Drolet, Firas Al-Hafez, Aditya Bhatt et al.

Discrete latent bottlenecks in variational autoencoders (VAEs) offer high bit efficiency and can be modeled with autoregressive discrete distributions, enabling parameter-efficient multimodal search with transformers. However, discrete random variables do not allow for exact differentiable parameterization; therefore, discrete VAEs typically rely on approximations, such as Gumbel-Softmax reparameterization or straight-through gradient estimates, or employ high-variance gradient-free methods such as REINFORCE that have had limited success on high-dimensional tasks such as image reconstruction. Inspired by popular techniques in policy search, we propose a training framework for discrete VAEs that leverages the natural gradient of a non-parametric encoder to update the parametric encoder without requiring reparameterization. Our method, combined with automatic step size adaptation and a transformer-based encoder, scales to challenging datasets such as ImageNet and outperforms both approximate reparameterization methods and quantization-based discrete autoencoders in reconstructing high-dimensional data from compact latent spaces, achieving a 20% improvement on FID Score for ImageNet 256.

Aditya Bhatt, Adrian Sieler, Steffen Puhlmann et al.

We present in-hand manipulation skills on a dexterous, compliant, anthropomorphic hand. Even though these skills were derived in a simplistic manner, they exhibit surprising robustness to variations in shape, size, weight, and placement of the manipulated object. They are also very insensitive to variation of execution speeds, ranging from highly dynamic to quasi-static. The robustness of the skills leads to compositional properties that enable extended and robust manipulation programs. To explain the surprising robustness of the in-hand manipulation skills, we performed a detailed, empirical analysis of the skills' performance. From this analysis, we identify three principles for skill design: 1) Exploiting the hardware's innate ability to drive hard-to-model contact dynamics. 2) Taking actions to constrain these interactions, funneling the system into a narrow set of possibilities. 3) Composing such action sequences into complex manipulation programs. We believe that these principles constitute an important foundation for robust robotic in-hand manipulation, and possibly for manipulation in general.

Aditya Bhatt, Daniel Palenicek, Boris Belousov et al.

Sample efficiency is a crucial problem in deep reinforcement learning. Recent algorithms, such as REDQ and DroQ, found a way to improve the sample efficiency by increasing the update-to-data (UTD) ratio to 20 gradient update steps on the critic per environment sample. However, this comes at the expense of a greatly increased computational cost. To reduce this computational burden, we introduce CrossQ: A lightweight algorithm for continuous control tasks that makes careful use of Batch Normalization and removes target networks to surpass the current state-of-the-art in sample efficiency while maintaining a low UTD ratio of 1. Notably, CrossQ does not rely on advanced bias-reduction schemes used in current methods. CrossQ's contributions are threefold: (1) it matches or surpasses current state-of-the-art methods in terms of sample efficiency, (2) it substantially reduces the computational cost compared to REDQ and DroQ, (3) it is easy to implement, requiring just a few lines of code on top of SAC.

Łukasz Kidziński, Carmichael Ong, Sharada Prasanna Mohanty et al.

In the NeurIPS 2018 Artificial Intelligence for Prosthetics challenge, participants were tasked with building a controller for a musculoskeletal model with a goal of matching a given time-varying velocity vector. Top participants were invited to describe their algorithms. In this work, we describe the challenge and present thirteen solutions that used deep reinforcement learning approaches. Many solutions use similar relaxations and heuristics, such as reward shaping, frame skipping, discretization of the action space, symmetry, and policy blending. However, each team implemented different modifications of the known algorithms by, for example, dividing the task into subtasks, learning low-level control, or by incorporating expert knowledge and using imitation learning.