Alexander C. Shkolnik

Removing Some ‘A’ from AI: Embodied Cultured Networks

We embodied networks of cultured biological neurons in simulation and in robotics. This is a new research paradigm to study learning, memory, and information processing in real time: the Neurally-Controlled Animat. Neural activity was subject to detailed electrical and optical observation using multi-electrode arrays and microscopy in order to access the neural correlates of animat behavior. Neurobiology has given inspiration to AI since the advent of the perceptron and consequent artificial neural networks, developed using local properties of individual neurons. We wish to continue this trend by studying the network processing of ensembles of living neurons that lead to higher-level cognition and intelligent behavior.

Bounding on rough terrain with the LittleDog robot

A motion planning algorithm is described for bounding over rough terrain with the LittleDog robot. Unlike walking gaits, bounding is highly dynamic and cannot be planned with quasi-steady approximations. LittleDog is modeled as a planar five-link system, with a 16-dimensional state space; computing a plan over rough terrain in this high-dimensional state space that respects the kinodynamic constraints due to underactuation and motor limits is extremely challenging. Rapidly Exploring Random Trees (RRTs) are known for fast kinematic path planning in high-dimensional configuration spaces in the presence of obstacles, but search efficiency degrades rapidly with the addition of challenging dynamics. A computationally tractable planner for bounding was developed by modifying the RRT algorithm by using: (1) motion primitives to reduce the dimensionality of the problem; (2) Reachability Guidance, which dynamically changes the sampling distribution and distance metric to address differential constraints and discontinuous motion primitive dynamics; and (3) sampling with a Voronoi bias in a lower-dimensional “task space” for bounding. Short trajectories were demonstrated to work on the robot, however open-loop bounding is inherently unstable. A feedback controller based on transverse linearization was implemented, and shown in simulation to stabilize perturbations in the presence of noise and time delays.

Inverse Kinematics for a Point-Foot Quadruped Robot with Dynamic Redundancy Resolution

In this work we examine the control of center of mass and swing leg trajectories in LittleDog, a point-foot quadruped robot. It is not clear how to formulate a function to compute forward kinematics of the center of mass of the robot as a function of actuated joint angles because point-foot walkers have no direct actuation between the feet and the ground. Nevertheless, we show that a whole-body Jacobian exists and is well defined when at least three of the feet are on the ground. Also, the typical approach of work-space centering for redundancy resolution causes destabilizing motions when executing fast motions. An alternative redundancy resolution optimization is proposed which projects single-leg inverse kinematic solutions into the nullspace. This hybrid approach seems to minimize 1) unnecessary rotation of the body, 2) twisting of the stance legs, and 3) whole-body involvement in achieving a step leg trajectory. In simulation, this control allows the robot to perform significantly more dynamic behaviors while maintaining stability.

Asymptotically-optimal path planning for manipulation using incremental sampling-based algorithms

A desirable property of path planning for robotic manipulation is the ability to identify solutions in a sufficiently short amount of time to be usable. This is particularly challenging for the manipulation problem due to the need to plan over high-dimensional configuration spaces and to perform computationally expensive collision checking procedures. Consequently, existing planners take steps to achieve desired solution times at the cost of low quality solutions. This paper presents a planning algorithm that overcomes these difficulties by augmenting the asymptotically-optimal RRT* with a sparse sampling procedure. With the addition of a collision checking procedure that leverages memoization, this approach has the benefit that it quickly identifies low-cost feasible trajectories and takes advantage of subsequent computation time to refine the solution towards an optimal one. We evaluate the algorithm through a series of Monte Carlo simulations of seven, twelve, and fourteen degree of freedom manipulation planning problems in a realistic simulation environment. The results indicate that the proposed approach provides significant improvements in the quality of both the initial solution and the final path, while incurring almost no computational overhead compared to the RRT algorithm. We conclude with a demonstration of our algorithm for single-arm and dual-arm planning on Willow Garages PR2 robot.

High-dimensional underactuated motion planning via task space control

Kinodynamic planning algorithms have the potential to find feasible control trajectories which accomplish a task even in very nonlinear or constrained dynamical systems. Underactuation represents a particular form of a dynamic constraint, inherently present in many machines of interest (e.g., walking robots), and necessitates planning for long-term control solutions. A major limitation in motion planning techniques, especially for real-time implementation, is that they are only practical for relatively low degree-of-freedom problems. Here we present a model-based dimensionality reduction technique based on an extension of partial feedback linearization control into a task-space framework. This allows one to plan motions for a complex underactuated robot directly in a low-dimensional task-space, and to resolve redundancy with lower-priority tasks. We illustrate the potential of this approach with an extremely simple motion planning system which solves the swing-up problem for multi-link underactuated pendula, and discuss extensions to the control of walking.

Collision detection in legged locomotion using supervised learning

We propose a fast approach for detecting collision- free swing-foot trajectories for legged locomotion over extreme terrains. Instead of simulating the swing trajectories and checking for collisions along them, our approach uses machine learning techniques to predict whether a swing trajectory is collision-free. Using a set of local terrain features, we apply supervised learning to train a classifier to predict collisions. Both in simulation and on a real quadruped platform, our results show that our classifiers can improve the accuracy of collision detection compared to a real-time geometric approach without significantly increasing the computation time.

Rotary High Efficiency Hybrid Cycle Engine

In this paper we discuss a rotary implementation of the High Efficiency Hybrid Cycle (HEHC) engine. HEHC is a thermodynamic cycle which borrows elements of Diesel, Otto and Atkinson cycles, characterized by 1) compression of air only (e.g. Diesel), 2) constant volume heat addition (e.g. Otto), and 3) expansion to atmospheric pressure (e.g. Atkinson). The engine consists of a compressor, an isolated combustion chamber, and an expander. Both compressor and expander consist of a simple design with two main parts: a rotor and an oscillating rocker. Compared to conventional internal combustion engines, in which all processes happen within the same space but at different times, in this engine, all processes are occurring simultaneously but in different chambers, allowing for independent optimization of each process. The result is an engine which may offer up to 57% peak efficiency, and above 50% sustained efficiency across typical driving loads.

Development of a Small Rotary SI/CI Combustion Engine

This paper describes the development of small rotary internal combustion engines developed to operate on the High Efficiency Hybrid Cycle (HEHC). The cycle, which combines high compression ratio (CR), constant-volume (isochoric) combustion, and overexpansion, has a theoretical efficiency of 75% using air-standard assumptions and first-law analysis. This innovative rotary engine architecture shows a potential indicated efficiency of 60% and brake efficiency of >50%. As this engine does not have poppet valves and the gas is fully expanded before the exhaust stroke starts, the engine has potential to be quiet. Similar to the Wankel rotary engine, the ‘X’ engine has only two primary moving parts – a shaft and rotor, resulting in compact size and offering low-vibration operation. Unlike the Wankel, however, the X engine is uniquely configured to adopt the HEHC cycle and its associated efficiency and low-noise benefits. The result is an engine which is compact, lightweight, low-vibration, quiet, and fuel-efficient. Two prototype engines are discussed. The first engine is the larger X1 engine (70hp), which operates on the HEHC with compression-ignition (CI) of diesel fuel. A second engine, the XMv3, is a scaled down X engine (70cc / 3HP) which operates with spark-ignition (SI) of gasoline fuel. Scaling down the engine presented unique challenges, but many of the important features of the X engine and HEHC cycle were captured. Preliminary experimental results including firing analysis are presented for both engines. Further tuning and optimization is currently underway to fully exploit the advantages of HEHC with the X architecture engines.

High Efficiency Hybrid Cycle Engine

A “High Efficiency Hybrid Cycle” (HEHC) thermodynamic cycle is explored. This four-stroke cycle borrows elements from Otto, Diesel, Atkinson, and Rankine cycles. Air is compressed into an isolated combustion chamber, allowing for true isochoric combustion, and extended duration for combustion to proceed until completion. Combustion products expand into a chamber with greater volume than intake. We provide details of a compact HEHC design implementation using rotary pistons and isolated rotating combustion chambers. Two Pistons simultaneously rotate and reciprocate and are held in position by two roller bearings. One Piston performs intake and compression, while the other performs exhaust and expansion. We predict a reduction of energy losses, moving part counts, weight and size over conventional engines.Copyright

Performance of a Low-Blowby Sealing System for a High Efficiency Rotary Engine

The X engine is a non-Wankel rotary engine that allies high power density and high efficiency by running a high-pressure Atkinson cycle at high speeds. The X engine overcomes the gas leakage issue of the Wankel engine by using two axially-loaded face seals that directly interface with three stationary radially-loaded apex seals per rotor. The direct-interfacing of the apex and face seals eliminates the need for corner seals of the typical Wankel engine, significantly reducing rotary engine blowby. This paper demonstrates the sealing performance that can be achieved by this new type of seal configuration for a rotary engine based on dynamics models and experiments. The dynamics models calculate the displacement and deformation of the face and apex seals for every crank angle using a time implicit solver. The gas leakage is then calculated from the position of the seals and pressure in the chambers and integrated over a rotor revolution. An “effective leakage orifice” area can be determined, to compare blowby between different engine types. Model results show that the X engine equivalent leakage area could be around 35% that of the leakage area of a similarly sized Wankel engine obtained from the same modeling method, which brings the X engine leakage closer to the piston engine’s leakage range. Initial experimental results support the findings from the model, as the X engine shows an equivalent leakage area of about 65% that of a scaled Wankel engine. This result demonstrates the potential of the X engine to achieve gas sealing improvements through additional seal development. Introduction For applications that need a high power density, the rotary engine has been an interesting candidate since its debut in the 1950s. In addition to its impressive power density, it also features fewer moving parts and lower vibrations levels compared to piston engines. However, it has seen a decrease in popularity in recent years, notably in the automotive industry, with the last remaining Wankel engine powered car being manufactured in 2012. The withdrawal of rotary engines from the automotive market can be explained by the traditional drawbacks of rotary engines, along with increasing emission regulations around the globe. An important drawback of the rotary engines is the difficulty to seal the combustion chamber. The Wankel engine’s geometry requires radially loaded apex seals as well as axially loaded face seals, and the interface between these is not effectively sealed which leads to increased leakage. Various scientific work has been done on the Wankel type rotary engines since the 1950’s. The sealing performance of the Wankel engine was studied both experimentally and by modeling. Different methods were used to simulate the Wankel engine by the past, such as modifying piston engines commercial simulation softwares [1], by the use of CFD tools [2], or by analytical calculations [3, 4]. Throughout the literature, the seal leakage values are usually quantified as an equivalent orifice area, which is not directly linked to the dynamics of the seals. This leakage area is usually determined by fitting a full engine cycle simulation model to experimental in-cylinder pressure and other data, tuning parameters such as compression ratio, port flow coefficients, heat transfer multipliers, and leakage orifice areas. For a Mazda Wankel, Eberle and Klomp [5] determined a leakage area value of 2 mm2 per cell, while others [6]-[8] found 1 mm2. In an effort to better understand the behavior of the Wankel seals, models were created to study the dynamics of the seals [9]-[12]. The seal model by Picard [3, 4] successfully relates the sealing performance to the dynamics and deformation of the seals to predict leakage. The final results of the model predict an equivalent leakage area varying between 1 mm2 and 2 mm2 for the Renesis engine. The main conclusion of these studies is that the seals tend to leak near the extremities of the parts that interact with each other. In a piston engine, since the rings only have a single gap, they are able to achieve better sealing performances than Wankel engines. The X Engine is a rotary engine with an alternative architecture that has the potential to solve the rotary engine sealing Downloaded from SAE International by Alexander Shkolnik, Sunday, May 27, 2018