Yearly Recap: Robotics & AI 2023

Paradigm
Paradigm
Published in
13 min readDec 28, 2023

The robotics and artificial intelligence (AI) landscape witnessed a dynamic and transformative year, marked by noteworthy developments, challenges, and breakthroughs. From the inception of ultrasonically actuated glass needles enabling precise liquid handling to the creation of ‘brainless’ soft robots navigating complex environments, the year showcased the versatility and adaptability of robotic systems. Advancements extended to diverse domains, including the integration of learning AI systems mimicking human voices after brief recordings and the development of resilient artificial muscles empowering insect-scale aerial robots to recover from severe damage. Innovations also spanned soft robotic tentacles for medical procedures, caterpillar-inspired robots adept at confined space navigation, and underwater robots designed for ocean waste collection. The year also witnessed the design of a quadrupedal robot climbing steps and navigating uneven environments without visual or tactile sensors, emphasizing innovation in challenging scenarios.

This recap captures the rapid evolution of robotic systems and AI technologies, delving into breakthroughs in machine learning algorithms, the proliferation of collaborative robots, and the integration of AI into various facets of automation. These innovations are reshaping manufacturing, healthcare, finance, and logistics industries, paving the way for increased efficiency, productivity, and new possibilities.

Let’s dive into some key highlights of the year:

January — February

Researchers have developed an ultrasonically actuated glass needle that can be attached to a robotic arm. This lets them pump and mix minuscule amounts of liquid and trap particles.

The research team integrated deep-learning techniques with the use of drones to automatically detect defects on the reflector surface. Specifically, they began by manually controlling a drone equipped with a high-resolution RGB camera to fly over the surface along a predetermined route.

Researchers have developed a new reconfigurable workspace soft (RWS) robotic gripper that can scoop, pick and grasp a wide range of consumer items. The RWS gripper’s comprehensive and adaptive capabilities make it particularly useful in logistics and food industries where they depend on robotic automation to meet increasing demands in efficiently picking and packing items.

The research team established a multi-domain framework for switched electromechanical dynamics. Until then, researchers seeking to predict the energy usage of robotic systems were forced to rely on a piecemeal method providing only rough approximations under limited conditions.

A team of researchers at Microsoft has demonstrated a new AI system that is capable of mimicking a person’s voice after training with a recording just three seconds long.

— Inspired by sea cucumbers, engineers have designed miniature robots that rapidly and reversibly shift between liquid and solid states. On top of being able to shape-shift, the robots are magnetic and can conduct electricity.

Researchers have made a significant leap forward in developing insect-sized jumping robots capable of performing tasks in the small spaces often found in mechanical, agricultural and search-and-rescue settings.

Researchers have developed a new continuum robot inspired by the trunks of elephants. This robot has a customizable design that allows it to be tailored for different applications.

Researchers have developed a new framework that could provide four-legged robots with leader-following abilities in both nighttime and daytime conditions. This framework is based on visual and LiDAR detection technology.

Researchers developed a new approach to building deformable underwater robots, using simple repeating substructures.

A research group has developed a learning method for AI that combines classification performance for data with multiple labels with the ability to learn continually from data. Numerical experiments on real-world multi-label data indicate that the new method outperforms conventional approaches.

Researchers tested pigeons’ learning abilities and concluded the birds employ the same basic process, called associative learning, as the most advanced AI technologies.

— Engineers have developed AI technology to predict if women with breast cancer would benefit from chemotherapy prior to surgery.

Researchers have developed a soft robotic tentacle that could be used to improve the efficiency of some standard medical procedures. This tentacle, is controlled through their novel control algorithm, together with the so-called active cooling for shape memory alloy, the actuating candidate for the robot.

March — May

Researchers have created soft robots that can seamlessly shift from walking to swimming, for example, or crawling to rolling using a bistable actuator made of 3D-printed soft rubber containing shape-memory alloy springs that react to electrical currents by contracting, which causes the actuator to bend.

Researchers have developed resilient artificial muscles that can enable insect-scale aerial robots to effectively recover flight performance after suffering severe damage.

Engineers have now developed a system that would enable autonomous vehicles to explore caves, lava tubes and even oceans on other worlds on their own.

Researchers created RoboNinja, a machine learning-based system that could allow robots to cut multi-material objects, particularly soft objects with hard cores.

Researchers created Real-time Neural MPC, a framework that integrates complex model architectures based on artificial neural networks (ANNs) in an MPC framework for agile robots (i.e., quadrotors — drones with four rotors).

Scientists have developed fully biodegradable, high-performance artificial muscles. Their research project marks another step towards green technology becoming a lasting trend in the field of soft robotics.

A research study proves that AI can respond to complex survey questions just like a real human.

Researchers found that preschoolers prefer learning from what they perceive as a competent robot over an incompetent human. This study is the first to use both a human speaker and a robot to see if children deem social affiliation and similarity more important than competency when choosing which source to trust and learn from.

Researchers have demonstrated a caterpillar-like soft robot that can move forward, backward and dip under narrow spaces.

Researchers have developed biosensor technology that will allow you to operate devices, such as robots and machines, solely through thought control.

— A multidisciplinary team has created a new fabrication technique for fully foldable robots that can perform a variety of complex tasks without relying on semiconductors.

Researchers developed a novel robotic link with tactile and proximity sensing capabilities. Additionally, they created a simulation and learning framework that can be employed to train the robotic link to sense its environment.

Researchers have designed a low-cost, energy-efficient robotic hand that can grasp a range of objects using just the movement of its wrist and the feeling in its ‘skin’.

Roboticists have developed a jellyfish-inspired underwater robot with which they hope one day to collect waste from the bottom of the ocean. The almost noise-free prototype trap objects underneath its body without physical contact, thereby enabling safe interactions in delicate environments such as coral reefs.

Researchers have developed a robot, called ReMotion, that occupies physical space on a remote user’s behalf, automatically mirroring the user’s movements in real time and conveying key body language that is lost in standard virtual environments.

June — August

Researchers have established a new approach for additively manufacturing soft robotics, using a 3D knitting method that can holistically “print” entire soft robots.

Researchers at Stanford University have developed digital skin that can convert sensations such as heat and pressure to electrical signals that can be read by electrodes implanted in the human brain.

Researchers used novel approach to teach robot to navigate over obstacles. They called their approach ViNL (Visual Navigation and Locomotion).

A team of engineering researchers has developed a quadrupedal robot technology that can climb up and down the steps and moves without falling over in uneven environments such as tree roots without the help of visual or tactile sensors even in disastrous situations in which visual confirmation is impeded due to darkness or thick smoke from the flames.

Researchers developed a worm-inspired robot with a body structure that is based on the oriental paper-folding art of origami. This robotic system is based on actuators that respond to magnetic forces, compressing and bending its body to replicate the movements of worms.

FluidLab, a simulation tool from researchers at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), enhanced robot learning for complex fluid manipulation tasks like making latte art, ice cream, and even manipulating air.

MIT engineers have introduced a framework for how remote human supervision could be scaled to make a hybrid system efficient without compromising passenger safety.

A research team introduced NerfBridge, a new open-source software package for training NeRF algorithms that could ultimately enable their use in online robotics experiments.

A team of researchers has developed a new method for controlling lower limb exoskeletons using deep reinforcement learning. The method enables more robust and natural walking control for users of lower limb exoskeletons.

Engineers have developed a new model that trains four-legged robots to see more clearly in 3D. The advance enabled a robot to autonomously cross challenging terrain with ease — including stairs, rocky ground and gap-filled paths — while clearing obstacles in its way.

Scientists have developed a model that utilizes AI to accurately classify cardiac functions and valvular heart diseases from chest radiographs.

— A research team has developed ROSE, a novel embracing soft gripper inspired by the blooming and closing of rose flowers. Bearing a surprisingly simple, inexpensive, and scalable design, ROSE can pick up many kinds of objects without damaging them, even in challenging environments and conditions.

— Researchers have made advancements in bionics with the development of a new electric variable-stiffness artificial muscle. The innovative technology possesses self-sensing capabilities and has the potential to revolutionize soft robotics and medical applications.

— Engineers inspired by the morphing skins of animals like chameleons and octopuses have developed capillary-controlled robotic flapping fins to create switchable optical and infrared light multipixel displays that are 1,000 times more energy efficient than light-emitting devices.

Researchers have trained a robotic ‘chef’ to watch and learn from cooking videos and recreate the dish itself.

Researchers have developed an L3 F-TOUCH sensor to enhance tactile capabilities in robots, allowing it to ‘feel’ objects and adjust its grip accordingly.

Researchers have developed a new explainable AI model to reduce bias and enhance trust and accuracy in machine learning-generated decision-making and knowledge organization.

The research team created a noninvasive electroencephalogram (EEG) sensor that they installed in a Meta VR headset that can be worn comfortably for long periods.

September — December

An innovative bimanual robot displays tactile sensitivity close to human-level dexterity using AI to inform its actions.

Engineers have developed HADAR, or heat-assisted detection and ranging.

A new AI technique enables a robot to develop complex plans for manipulating an object using its entire hand, not just fingertips.

Researchers have created a ‘brainless’ soft robot that can navigate more complex and dynamic environments.

A research team has developed ‘soft valve’ technology — an all-in-one solution that integrates sensors and control valves while maintaining complete softness.

Researchers have developed small robotic devices that can change how they move through the air by ‘snapping’ into a folded position during their descent.

A team of scientists developed a machine learning model for discovering critical-element-free permanent magnet materials based on the predicted Curie temperature of new material combinations.

A team has developed a shape-changing smart speaker, which uses self-deploying microphones to divide rooms into speech zones and track the positions of individual speakers.

Scientists have created a non-invasive movement tracking method called GlowTrack that uses fluorescent dye markers to train artificial intelligence to capture movement, from a single mouse digit to the human hand.

A study shows that human instruction is still necessary to detect and compensate for unintended, and sometimes negative, changes in neurosurgeon behavior after virtual reality AI training.

Researchers combined soft microactuators with high-energy-density chemical fuel to create an insect-scale quadrupedal robot that is powered by combustion and can outrace, outlift, outflex and outleap its electric-driven competitors.

A study has found that people can learn to use supernumerary robotic arms as effectively as working with a partner in just one hour of training.

— Researchers have designed a robot which can change form to tackle varying scenarios.

Researchers developed AI that can intelligently design robots from scratch by compressing billions of years of evolution into mere seconds. It’s not only fast but also runs on a lightweight computer and designs wholly novel structures from scratch — without human-labeled, bias-filled datasets.

— Engineers created a 2-centimeter-long modular robot called mCLARI that can passively change shape to navigate narrow gaps.

Researchers introduced a new strategy for constructing 3D force sensors using knots and optical fibers, improving pressure, friction, and slip sensing performance.

— Researchers developed Feature Fields for Robotic Manipulation (F3RM), enabling robots to identify and grasp objects in complex environments based on 2D images and 3D features.

— Researchers developed a soft, stretchable, and highly sensitive sensor skin for robotics and prosthetics, enabling touch sensitivity and dexterity.

— A study showed that pigeons approach problem-solving similarly to artificial intelligence, helping them tackle complex tasks.

— Researchers have succeeded in printing a robotic hand with bones, ligaments and tendons made of different polymers using a new laser scanning technique. The new technology makes it possible to 3D print special plastics with elastic qualities in one go.

2023 Robotics Awards Winners

IERA Award

ArmMotusTM EMU (Fourier Intelligence): ArmMotus™ EMU is a 3D back-drivable upper limb rehabilitation robot that adopted an innovative cable-driven mechanism, combined with a parallel structure made of lightweight carbon fibre rods which perfectly reduces the friction and inertia of the device. This enables the control system to respond and execute more efficiently, resulting in higher compliance in human-machine interaction.

Robotics Innovation Awards

List of Organizations and Innovations:

ABB — ABB introduces its smallest robot arm ever

Agility Robotics — Agility’s Digit humanoid targets logistics tasks

Alert Innovation — Alert Innovation acquired by Walmart to boost fulfillment

Amazon — Amazon looks to automate ‘Holy Grail’ with Sparrow robot arm

AMD — AMD Kria KR260 platform eases robotics development

American Robotics — American Robotics expands drone services to oil and gas industries

Australian Droid + Robot — Robot explores dangerous caves to assess vulnerable bat colonies

Boston Dynamics — Boston Dynamics Stretch finds early success unloading trailers

Brain Corp — Brain Corp adds dual-purpose to floor-scrubbing robots

Cionic — Cionic bionic sleeve assists people with mobility issues

Clearpath Robotics — Clearpath OutdoorNav software eases mobile robotics development

Coalition Against Weaponized Robots — Coalition takes stand against weaponized robots

Deepmind — Deepmind open-sources MujoCo physics engine

Foxglove — Foxglove standardizes robotics data recording with MCAP

GelSight — MIT spinoff creates affordable, mini tactile sensor

HEBI Robotics — HEBI applies modularity expertise to mobile robots

Inbolt — Inbolt simplifies the use of vision servoing with a unique application

Intrinsic — Intrinsic acquires Open Source Robotics Corp

Inuitive — Inuitive’s vision technologies improve robot autonomy

James Webb Space Telescope — James Webb Space Telescope revolutionizes astronomy

Johns Hopkins University — NASA succeeds in 1st planetary defense test mission

Locus Robotics — Locus Robotics surpasses 1 billion picks with AMRs

Machina Labs — Machina Labs uses robots to manufacture titanium sheet metal parts

Main Street Autonomy — Main Street Autonomy simplifies sensor calibration

MassRobotics — MassRobotics aims to diversify robotics talent pool

Meijer — Meijer uses robots to clean up Great Lakes

NASA JPL — Perseverance Rover completes first multi-sol drive on Mars

Nauticus Robotics — Nauticus dives into future of underwater robotics

NVIDIA — NVIDIA improves ROS 2 pipeline processing with hardware acceleration

OnRobot — OnRobot D:PLOY lowers barriers to robotics adoption

Open-Source Leg Project — Open-source leg project democratizes prosthetic development

Ottonomy — Yeti delivery robot autonomously dispenses packages

Photoneo — MotionCam-3D Color camera captures dynamic scenes

PickNik Robotics — MoveIt Studio offers low-code robotics development environment

Rapid Robotics — Rapid Robotics brings flexibility to robotic workcells

RoboTire — RoboTire transforms tire-changing process

Skyline Robotics — Skyline Robotics disrupting B window cleaning industry

Spacee — Spacee Deming offers simplified approach to shelf-scanning robots

SparkAI — SparkAI helps John Deere resolve agriculture autonomy challenges

Stratom — Stratom makes vehicle refueling safer, more efficient

Tatum Robotics — Tatum creates revolutionary communication tool for deaf-blind people

Telexistence — Telexistence deploys restocking robots at Japanese convenience stores

Teradyne — Teradyne merges AMR subsidiaries AutoGuide & MIR

Tuskrobots — Tuskrobots develops novel AMR/AGV to automate pallet movement

UC Berkeley — DayDreamer algorithm enables robot learning without simulation

Universal Robots — UR20 is the company’s fastest, strongest cobot ever

Viam Robotics — Viam combines open-source robotics system with integrated cloud services

Waymo — Waymo continues robotaxi expansion

WHILL — WHILL autonomously transports passengers at 1st North American airport

WiBotic — WiBotic brings the benefits of wireless charging technology to ASRS

In the dynamic realm of robotics, the past year unfolded with a cascade of breakthroughs that spanned diverse domains. Researchers showcased their ingenuity, introducing innovations ranging from ultrasonically actuated glass needles, providing precision in liquid handling, to the development of a learning AI system adept at mimicking human voices after a brief three-second recording. The boundaries of robotic capabilities were continually pushed, evident in the creation of soft robotic tentacles designed for medical procedures and caterpillar-like robots displaying expertise in navigating confined spaces. Additionally, resilient artificial muscles emerged, empowering insect-scale aerial robots to recover effectively from severe damage.

The integration of AI permeated various applications, from predicting the efficacy of chemotherapy for breast cancer patients to enhancing robots’ proficiency in cutting multi-material objects. Notable breakthroughs also featured the introduction of an underwater robot inspired by jellyfish, engineered for the collection of ocean waste, and the creation of a ‘brainless’ soft robot adept at navigating complex environments. As researchers explored new frontiers, 3D printing techniques for fully foldable robots gained prominence, alongside the introduction of a framework facilitating remote human supervision in autonomous systems. The year showcased a vibrant tapestry of advancements, reinforcing the ever-expanding horizons within the field of robotics and AI.

Main sources

Research articles

Science Robotics

Science Daily

IEEE Spectrum

--

--