-
Automated Physics-Derived Code Generation for Sensor Fusion and State Estimation
Authors:
Orestis Kaparounakis,
Vasileios Tsoutsouras,
Dimitrios Soudris,
Phillip Stanley-Marbell
Abstract:
We present a new method for automatically generating the implementation of state-estimation algorithms from a machine-readable specification of the physics of a sensing system and physics of its signals and signal constraints. We implement the new state-estimator code generation method as a backend for a physics specification language and we apply the backend to generate complete C code implementa…
▽ More
We present a new method for automatically generating the implementation of state-estimation algorithms from a machine-readable specification of the physics of a sensing system and physics of its signals and signal constraints. We implement the new state-estimator code generation method as a backend for a physics specification language and we apply the backend to generate complete C code implementations of state estimators for both linear systems (Kalman filters) and non-linear systems (extended Kalman filters). The state estimator code generation from physics specification is completely automated and requires no manual intervention. The generated filters can incorporate an Automatic Differentiation technique which combines function evaluation and differentiation in a single process. Using the description of physical system of a range of complexities, we generate extended Kalman filters, which we evaluate in terms of prediction accuracy using simulation traces. The results show that our automatically-generated sensor fusion and state estimation implementations provide state estimation within the same error bound as the human-written hand-optimized counterparts. We additionally quantify the code size and dynamic instruction count requirements of the generated state estimator implementations on the RISC-V architecture. The results show that our synthesized state estimation implementation employing Automatic Differentiation leads to an average improvement in the dynamic instruction count of the generated Kalman filter of 7%-16% compared to the standard differentiation technique. This is improvement comes at the limited cost of an average 4.5% increase in the code size of the generated filters.
△ Less
Submitted 28 April, 2020;
originally announced April 2020.
-
Synthesizing Compact Hardware for Accelerating Inference from Physical Signals in Sensors
Authors:
Vasileios Tsoutsouras,
Max Vigdorchik,
Phillip Stanley-Marbell
Abstract:
We present dimensional circuit synthesis, a new method for generating digital logic circuits that improve the efficiency of training and inference of machine learning models from sensor data. The hardware accelerators that the method generates are compact enough (a few thousand gates) to allow integration within low-cost miniaturized sensor integrated circuits, right next to the sensor transducer.…
▽ More
We present dimensional circuit synthesis, a new method for generating digital logic circuits that improve the efficiency of training and inference of machine learning models from sensor data. The hardware accelerators that the method generates are compact enough (a few thousand gates) to allow integration within low-cost miniaturized sensor integrated circuits, right next to the sensor transducer. The method takes as input a description of physical properties of relevant signals in the sensor transduction process and generates as output a Verilog register transfer level (RTL) description for a circuit that computes low-level features that exploit the units of measure of the signals in the system.
We implement dimensional circuit synthesis as a backend to the compiler for Newton, a language for describing physical systems. We evaluate the backend implementation and the hardware it generates, on descriptions of 7 physical systems. The results show that our implementation of dimensional circuit synthesis generates circuits of as little as 1662 logic cells / 1239 gates for the systems we evaluate.
We synthesize the designs generated by the dimensional circuit synthesis compilation backend for a low-power miniature FPGA targeted by its manufacturer at sensor interface applications. The circuits which the method generated use as little as 27% of the resources of the 2.15x2.5 mm FPGA. We measure the power dissipation of the FPGA's isolated core supply rail and show that, driven with a pseudorandom signal input stream, the synthesized designs use as little as 1.0 mW and no more than 5.8 mW. These results show the feasibility of integrating physics-inspired machine learning methods within low-cost miniaturized sensor integrated circuits, right next to the sensor transducer.
△ Less
Submitted 4 February, 2020;
originally announced February 2020.
-
Payload-Mass-Aware Trajectory Planning on Multi-User Autonomous Unmanned Aerial Vehicles
Authors:
Vasileios Tsoutsouras,
Joseph Story,
Phillip Stanley-Marbell
Abstract:
Future unmanned aerial vehicles (drones) will be shared by multiple users and will have to operate in conditions where their fully-autonomous function is required. Calculation of a drones trajectory will be important but optimal trajectories cannot be calculated unless mass and flight speed are taken into account. This article presents the case for on-drone trajectory planning in a multi-user dyna…
▽ More
Future unmanned aerial vehicles (drones) will be shared by multiple users and will have to operate in conditions where their fully-autonomous function is required. Calculation of a drones trajectory will be important but optimal trajectories cannot be calculated unless mass and flight speed are taken into account. This article presents the case for on-drone trajectory planning in a multi-user dynamic payload mass scenario, allowing a drone to calculate its trajectory with no need for ground control communication. We formulate and investigate on-drone trajectory planning under variable payload mass and flight speed awareness, in cases where it is shared by multiple users or applications. We present efficient solutions using a combination of heuristic and optimization algorithms. To support this investigation, we present a new model for the power dissipation of drone propulsion as a function of speed and payload mass. We evaluate our proposed algorithmic solution on contemporary embedded processors and demonstrate its capability to generate near-optimal trajectories with limited computational overhead (less than 300 milliseconds on an ARM Cortex-A9 SoC).
△ Less
Submitted 8 January, 2020;
originally announced January 2020.
