Machine Learning Based Path Planning for Improved Rover Navigation (Pre-Print Version)

TL;DR

This paper presents two heuristics that, given a terrain heightmap around the rover, produce cost estimates that more effectively rank the candidate paths before ACE evaluation, and uses a machine learning (ML) model to predict areas that will be deemed untraversable by ACE.

Abstract

Enhanced AutoNav (ENav), the baseline surface navigation software for NASA's Perseverance rover, sorts a list of candidate paths for the rover to traverse, then uses the Approximate Clearance Evaluation (ACE) algorithm to evaluate whether the most highly ranked paths are safe. ACE is crucial for maintaining the safety of the rover, but is computationally expensive. If the most promising candidates in the list of paths are all found to be infeasible, ENav must continue to search the list and run time-consuming ACE evaluations until a feasible path is found. In this paper, we present two heuristics that, given a terrain heightmap around the rover, produce cost estimates that more effectively rank the candidate paths before ACE evaluation. The first heuristic uses Sobel operators and convolution to incorporate the cost of traversing high-gradient terrain. The second heuristic uses a machine learning (ML) model to predict areas that will be deemed untraversable by ACE. We used physics simulations to collect training data for the ML model and to run Monte Carlo trials to quantify navigation performance across a variety of terrains with various slopes and rock distributions. Compared to ENav's baseline performance, integrating the heuristics can lead to a significant reduction in ACE evaluations and average computation time per planning cycle, increase path efficiency, and maintain or improve the rate of successful traverses. This strategy of targeting specific bottlenecks with ML while maintaining the original ACE safety checks provides an example of how ML can be infused into planetary science missions and other safety-critical software.

Figures (5)

• Figure 1: A view of the ENav simulation environment. The green-yellow terrain shows how the Gradient Convolution heuristic, developed in this work, has assessed the cost of traversing the terrain (yellow regions are higher cost) and steers the rover toward safer regions.
• Figure 2: Given a height map (a), Sobel operators can be used to create a gradient or gradient squared map (b). The gradient map can be convolved with a kernel representing the orientation-agnostic footprint of the rover (c) to form the gradient cost map (d), which estimates the cost to traverse a location within the map.
• Figure 3: An example of a Learned Heuristic. Sets of terrain heightmaps (left) and maps generated by the ACE algorithm (center) were used to train a neural network to generate an inferred ACE probability map (right).
• Figure 4: ROS software architecture for testing ENav against the HDSim physics simulator
• Figure 5: A summary of key rover path planning performance metrics across various experiments. Incorporating heuristics produced more efficient paths, reduced the number of costly ACE evaluations, and maintained or slightly increased the rate of successfully reaching the goal.