1r.walk(1) GRASS GIS User's Manual r.walk(1)
2
6 r.walk - Creates a raster map showing the anisotropic cumulative cost
7 of moving between different geographic locations on an input raster map
8 whose cell category values represent cost.
9
11 raster, cost surface, cumulative costs, cost allocation
12
14 r.walk
15 r.walk --help
16 r.walk [-knrib] elevation=name friction=name output=name [solver=name]
17 [nearest=name] [outdir=name] [start_points=name]
18 [stop_points=name] [start_raster=name] [start_coordi‐
19 nates=east,north[,east,north,...]] [stop_coordi‐
20 nates=east,north[,east,north,...]] [max_cost=value]
21 [null_cost=value] [memory=memory in MB] [walk_coeff=a,b,c,d]
22 [lambda=float] [slope_factor=float] [--overwrite] [--help]
23 [--verbose] [--quiet] [--ui]
24 Flags:
26 -k
27 Use the ’Knight’s move’; slower, but more accurate
28 -n
30 Keep null values in output raster map
31 -r
33 Start with values in raster map
34 -i
36 Print info about disk space and memory requirements and exit
37 -b
39 Create bitmask encoded directions
40 --overwrite
42 Allow output files to overwrite existing files
43 --help
45 Print usage summary
46 --verbose
48 Verbose module output
49 --quiet
51 Quiet module output
52 --ui
54 Force launching GUI dialog
55 Parameters:
57 elevation=name [required]
58 Name of input elevation raster map
59 friction=name [required]
61 Name of input raster map containing friction costs
62 output=name [required]
64 Name for output raster map to contain walking costs
65 solver=name
67 Name of input raster map solving equal costs
68 Helper variable to pick a direction if two directions have equal
69 cumulative costs (smaller is better)
70 nearest=name
72 Name for output raster map with nearest start point
73 outdir=name
75 Name for output raster map to contain movement directions
76 start_points=name
78 Name of starting vector points map
79 Or data source for direct OGR access
80 stop_points=name
82 Name of stopping vector points map
83 Or data source for direct OGR access
84 start_raster=name
86 Name of starting raster points map
87 start_coordinates=east,north[,east,north,...]
89 Coordinates of starting point(s) (E,N)
90 stop_coordinates=east,north[,east,north,...]
92 Coordinates of stopping point(s) (E,N)
93 max_cost=value
95 Maximum cumulative cost
96 Default: 0
97 null_cost=value
99 Cost assigned to null cells. By default, null cells are excluded
100 memory=memory in MB
102 Maximum memory to be used (in MB)
103 Cache size for raster rows
104 Default: 300
105 walk_coeff=a,b,c,d
107 Coefficients for walking energy formula parameters a,b,c,d
108 Default: 0.72,6.0,1.9998,-1.9998
109 lambda=float
111 Lambda coefficients for combining walking energy and friction cost
112 Default: 1.0
113 slope_factor=float
115 Slope factor determines travel energy cost per height step
116 Default: -0.2125
117
119 r.walk computes anisotropic cumulative cost of moving between different
120 geographic locations on an input elevation raster map whose cell cate‐
121 gory values represent elevation combined with an input raster map layer
122 whose cell values represent friction cost.
123 r.walk outputs 1) a raster map showing the lowest cumulative cost
125 (time) of moving between each cell and the user-specified starting
126 points and 2) a second raster map showing the movement direction to the
127 next cell on the path back to the start point (see Movement Direction).
128 It uses an input elevation raster map whose cell category values repre‐
129 sent elevation, combined with a second input raster map whose cell val‐
130 ues represent friction costs.
131 This function is similar to r.cost, but in addition to a friction map,
133 it considers an anisotropic travel time due to the different walking
134 speed associated with downhill and uphill movements.
135
137 The formula from Aitken 1977/Langmuir 1984 (based on Naismith’s rule
138 for walking times) has been used to estimate the cost parameters of
139 specific slope intervals:
140 T = a*delta_S + b*delta_H_uphill + c*delta_H_moderate_downhill + d*delta_H_steep_downhill
141 where:
142 • T is time of movement in seconds,
144 • delta S is the horizontal distance covered in meters,
146 • delta H is the altitude difference in meters.
148 The a, b, c, d walk_coeff parameters take in account movement speed in
150 the different conditions and are linked to:
151 • a: time in seconds it takes to walk for 1 meter a flat surface
153 (1/walking speed)
154 • b: additional walking time in seconds, per meter of elevation
156 gain on uphill slopes
157 • c: additional walking time in seconds, per meter of elevation
159 loss on moderate downhill slopes (use positive value for de‐
160 creasing cost)
161 • d: additional walking time in seconds, per meter of elevation
163 loss on steep downhill slopes (use negative value for increas‐
164 ing cost)
165 It has been proved that moving downhill is favourable up to a specific
166 slope value threshold, after that it becomes unfavourable. The default
167 slope value threshold (slope_factor) is -0.2125, corresponding to
168 tan(-12), calibrated on human behaviour (>5 and <12 degrees: moderate
169 downhill; >12 degrees: steep downhill). The default values for a, b, c,
170 d walk_coeff parameters are those proposed by Langmuir (0.72, 6.0,
171 1.9998, -1.9998), based on man walking effort in standard conditions.
172 The friction cost parameter represents a time penalty in seconds of ad‐
174 ditional walking time to cross 1 meter distance.
175 The lambda parameter is a dimensionless scaling factor of the friction
177 cost:
178 total cost = movement time cost + lambda * friction costs * delta_S
179 For a more accurate result, the "knight’s move" option can be used (al‐
181 though it is more time consuming). In the diagram below, the center lo‐
182 cation (O) represents a grid cell from which cumulative distances are
183 calculated. Those neighbours marked with an x are always considered for
184 cumulative cost updates. With the "knight’s move" option, the neigh‐
185 bours marked with a K are also considered.
186 K K
187 K x x x K
188 x O x
189 K x x x K
190 K K
191 The minimum cumulative costs are computed using Dijkstra’s algorithm,
193 that find an optimum solution (for more details see r.cost, that uses
194 the same algorithm).
195
197 The movement direction surface is created to record the sequence of
198 movements that created the cost accumulation surface. This movement di‐
199 rection surface can be used by r.path to recover a path from an end
200 point back to the start point. The direction of each cell points to‐
201 wards the next cell. The directions are recorded as degrees CCW from
202 East:
203 112.5 67.5 i.e. a cell with the value 135
204 157.5 135 90 45 22.5 means the next cell is to the north-west
205 180 x 360
206 202.5 225 270 315 337.5
207 247.5 292.5
208 Once r.walk computes the cumulative cost map as a linear combination of
210 friction cost (from friction map) and the altitude and distance covered
211 (from the digital elevation model), the associated movement direction
212 map can be used by r.path to find the minimum cost path.
213 r.walk, like most all GRASS raster programs, is also made to be run on
215 maps larger that can fit in available computer memory. As the algorithm
216 works through the dynamic list of cells it can move almost randomly
217 around the entire area. r.walk divides the entire area into a number of
218 pieces and swaps these pieces in and out of memory (to and from disk)
219 as needed. This provides a virtual memory approach optimally designed
220 for 2-D raster maps. The amount of memory to be used by r.walk can be
221 controlled with the memory option, default is 300 MB. For systems with
222 less memory this value will have to be set to a lower value.
223
225 We compute a map showing how far a lost person could get from the point
226 where he or she was last seen while taking into account the topography
227 and landcover.
228 g.region swwake_30m -p
229 # create friction map based on land cover
230 r.recode landclass96 out=friction rules=- << EOF
231 1:3:0.1:0.1
232 4:5:10.:10.
233 6:6:1000.0:1000.0
234 7:7:0.3:0.3
235 EOF
236 r.walk -k elevation=elev_ned_30m friction=friction output=walkcost \
237 start_coordinates=635576,216485 lambda=0.5 max=10000
238 # compute contours on the cost surface to better understand
239 # how far the person can get in certain time (1000 is in seconds)
240 r.contour walkcost output=walkcost step=1000
241
243 • Aitken, R. 1977. Wilderness areas in Scotland. Unpublished
244 Ph.D. thesis. University of Aberdeen.
245 • Steno Fontanari, University of Trento, Italy, Ingegneria per
247 l’Ambiente e il Territorio, 2000-2001. Svilluppo di metodolo‐
248 gie GIS per la determinazione dell’accessibilità territoriale
249 come supporto alle decisioni nella gestione ambientale.
250 • Langmuir, E. 1984. Mountaincraft and leadership. The Scottish
252 Sports Council/MLTB. Cordee, Leicester.
253
255 r.cost, r.path, r.in.ascii, r.mapcalc, r.recode, r.out.ascii
256
258 Based on r.cost written by :
259 Antony Awaida, Intelligent Engineering, Systems Laboratory, M.I.T.
260 James Westervelt, U.S.Army Construction Engineering Research Laboratory
261 Updated for Grass 5 by Pierre de Mouveaux (pmx@audiovu.com)
262 Initial version of r.walk:
264 Steno Fontanari, 2002
265 Current version of r.walk:
267 Franceschetti Simone, Sorrentino Diego, Mussi Fabiano and Pasolli Mat‐
268 tia
269 Correction by: Fontanari Steno, Napolitano Maurizio and Flor Roberto
270 In collaboration with: Franchi Matteo, Vaglia Beatrice, Bartucca Luisa,
271 Fava Valentina and Tolotti Mathias, 2004
272 Updated for GRASS 6.1:
274 Roberto Flor and Markus Neteler
275 Updated for GRASS GIS 7:
277 Markus Metz
278 Multiple path directions sponsored by mundialis
279
281 Available at: r.walk source code (history)
282 Accessed: Saturday Jan 21 21:15:29 2023
284 Main index | Raster index | Topics index | Keywords index | Graphical
286 index | Full index
287 © 2003-2023 GRASS Development Team, GRASS GIS 8.2.1 Reference Manual
289GRASS 8.2.1 r.walk(1)